Subscribe now

Let me start with a question.

When was the last time you typed something into Google and just got ten blue links back?

You probably can’t remember. And after this week, you’ll remember even less.

At its annual developer conference, Google announced what its VP of Search Elizabeth Reid called “the biggest upgrade to our Search box in over 25 years.” The search bar now expands to accommodate long, conversational queries, the way you’d talk to Claude or ChatGPT. Users can create personal search agents that monitor topics on their behalf, autonomously, over time. And a new Gemini mode called Spark runs in the background, continuously, even when your laptop is closed or your phone is locked.

Google wants to help you google less.

Read that again.

What Google actually announced

Four things matter for marketing leaders:

A new search box: Google calls it the biggest upgrade to the Search box in over 25 years. Built on Gemini 3.5 Flash, it dynamically expands to give you space to describe exactly what you need. It accepts text, images, files, videos, and open Chrome tabs as input. And it goes beyond autocomplete: AI helps you formulate your question before you even finish typing. Built for the questions that do not fit into keywords.

Search agents: Google is entering what it calls "the era of Search agents." Users can now create personal information agents that run in the background, 24/7, scanning the web, news, blogs, social posts, and real-time data sources for exactly what you asked for. The agent does not wait to be asked again. It monitors continuously and sends a synthesized update when something relevant happens. Looking for an apartment? Describe your criteria once and the agent scans listings for you. Want to know the moment a favorite athlete drops a sneaker collab? The agent notifies you instantly. The user does not search anymore. The agent searches.

Spark: Think of it as Google's answer to autonomous AI agents like OpenAI's Operator or Claude Cowork. A persistent background agent in Gemini that watches email inboxes, documents, and connected apps. It compiles, summarizes, and flags. Continuously. Without being asked. Crucially, it does not only start when a user gives a command. It triggers automatically when something happens, for example when a relevant email arrives in Gmail. MCP integrations are coming. It stays active even when your laptop is closed or your phone is locked.

Mini-apps and generative UI: Search can now build custom interactive tools on the fly, directly inside the search experience. Google calls this "agentic coding in Search." Ask for a fitness tracker that pulls in live maps, local weather, and real-time reviews, and Search codes it for you on the spot. For recurring tasks like planning a move or managing a project, Search builds persistent custom dashboards you can return to over time. These are shareable. Not just answers anymore. Fully functional, personalized interfaces, generated from a single query.

The number nobody expected

Here is the counterintuitive part.

You might expect that AI summaries are replacing searches. That people ask less because they get answers faster.

The opposite is true.

Google reported that AI Mode has surpassed one billion monthly users in just one year and that queries are more than doubling every quarter. Last quarter, total search queries reached an all-time high.

People are searching more than ever. They are just clicking on websites far less.

This is the paradox that defines the moment: more volume, less traffic. Google is growing. Your website is not necessarily benefiting.

Why this matters more than the demos

Google’s announcements were framed as convenience features for users.

They are actually a structural change to how brands get found.

In the old model, a customer had a question. They searched. They saw your ad or your link. They clicked. You had a chance.

In the new model, an agent has a task. It researches autonomously. It forms a shortlist. The human sees the conclusion, not the process.

👉 Your brand needs to be in the answer, not just in the auction.

This is not a future scenario. It is already happening. According to Pew Research data from 2025, when an AI summary appears in Google search, 26 percent of users end their browsing session entirely. They got what they needed. No click. No visit. No conversion path as it used to exist.

This week’s announcements accelerate that trend significantly.

The uncomfortable implication

If a Google agent is researching “best tools for X” on behalf of your potential customer, it is not looking at your creative. It is not impressed by your media budget.

It is reading what others say about you. It is checking third-party sources. It is evaluating whether your content actually answers the question, with specificity and credibility.

This is what I have been calling the shift from the Attention Economy to the Relevance Economy.

Attention you could buy. Relevance you have to earn.

The brands that will win in this environment are not the ones with the biggest ad budgets. They are the ones that consistently appear as the right answer in AI-generated summaries, in agent-compiled shortlists, in the outputs of systems that never saw a single impression of your campaign.

Subscribe now

Three questions for your next strategy session

• If a Google agent researches your product category today, does your brand appear in the answer?

• What percentage of your marketing budget is building the kind of credibility that AI systems can read and trust versus buying reach that AI agents skip?

• Do you even measure your presence in AI-generated answers right now?

If the answer to the last question is no, you are flying blind in the fastest-changing part of the marketing landscape.

Google’s CEO Sundar Pichai said this week that Google expects to spend between $180 and $190 billion this year on AI infrastructure. That is not a company hedging its bets.

That is a company betting everything.

The question is whether marketing leaders are updating their playbooks at the same speed.

Next week: the strategic framework for winning in the Relevance Economy. What it is, what the data shows, and what CMOs need to change.

Yours,

Prof. Dr. Andreas Fuchs 🦊🎓

👉 What was your reaction to Google I/O? I’d love to hear how other marketing leaders are thinking about this. Share in the comments.

Across boardrooms in 2026, marketing leaders are making the same bet at the same time.

They are investing record sums in AI. Content generation, predictive analytics, agentic SEO, dynamic creative. And they are doing it in the belief that this investment will set them apart from competitors who are doing the exact same thing, with the exact same tools, from the exact same vendors, often with strikingly similar prompts.

This is the quiet paradox of marketing in 2026. The technology every CMO is racing to adopt is, by definition, the one technology that cannot differentiate any of them.

And yet the strategic assumption underneath the entire AI-in-marketing conversation, that early and aggressive adoption will translate into competitive advantage, has gone almost completely unchallenged. It deserves to be challenged. Not with opinions, but with evidence.

The Thesis, in One Sentence

AI is becoming marketing infrastructure, not differentiation, and the strongest case for this comes from the founder of modern strategy theory himself.

In May 2025, Jay B. Barney, author of the most-cited paper in strategic management (Barney, 1991), published an article in MIT Sloan Management Review with a deliberately direct title: Why AI Will Not Provide Sustainable Competitive Advantage (Wingate et al., 2025).

Barney’s framework, taught in every MBA strategy course on earth, holds that a resource creates sustained advantage only when it is Valuable, Rare, Inimitable, and Organized (VRIO). His verdict on AI: valuable yes, rare no, inimitable no. “How can AI be the centerpiece of a sustained competitive advantage when everyone has it? We argue that it simply cannot” (Wingate et al., 2025).

Far from being a source of differentiation, AI is, in their words, a source of homogenization.

The Evidence for Commoditization

This is not a philosophical position. The data is brutal, and the collapse has only accelerated.

• Inference costs fell 280× in 18 months, and they have kept falling: Inference is what AI models do every time they answer a prompt. It is priced per million tokens, the unit of language AI processes (a million tokens is roughly 750,000 words, the length of a short novel). Stanford’s 2025 AI Index documented that running a model with GPT-3.5-level capability fell from $20 per million tokens in November 2022 to $0.07 by October 2024 (Maslej et al., 2025). The decline has not stopped. GPT-4 launched at $30 per million input tokens in March 2023. GPT-5 nano in 2026 lists at $0.05, a reduction of over 99% in three years. Stanford’s 2026 AI Index adds that hardware costs continue to fall 30% annually, and energy efficiency improves 40% annually (Maslej et al., 2026).

• Free models have caught up with paid ones: Open-weight models, AI models whose underlying code anyone can download and run, used to lag well behind closed commercial models from OpenAI, Anthropic, and Google. On Chatbot Arena, the leading head-to-head benchmark for AI quality, the performance gap between top open-weight and top closed-weight models shrank from 8.04% to 1.7% in a single year (Maslej et al., 2025).

• The same intelligence now runs on much smaller models: Two years ago, hitting 60% accuracy on MMLU, a standard test of general knowledge across 57 subjects, required a 540 billion parameter model. By 2024, a 3.8 billion parameter model cleared the same threshold. That is a 142× compression of the resources needed to achieve the same intelligence.

• And the gap between regions has nearly vanished: Three years ago, frontier AI development was almost exclusively American. By March 2026, the performance gap between the top US and top Chinese models had narrowed to just 2.7 percentage points (Maslej et al., 2026).

The translation for marketing leaders is simple. The AI capability your competitor uses today, you can buy at the same price tomorrow. What used to be a moat is now a utility, and it is getting cheaper faster than any technology in modern business history.

The Marketing-Specific Evidence

Marketing is where this homogenization hits hardest, and the peer-reviewed evidence is precise.

• AI raises individual creativity but lowers collective novelty. A Science Advances study by Doshi and Hauser (2024) found AI-assisted stories were 8.9 to 10.7% more similar to each other than human-written ones. The authors call it “an increase in individual creativity at the risk of losing collective novelty.” Marketing teams using the same models produce work that converges, even when each piece looks polished on its own.

• AI-generated content wins on search, which means more brands sound the same in the same places. A Marketing Science field experiment showed AI-generated SEO content is “virtually indistinguishable” from work by SEO experts, and outperforms human writers in search rankings (Reisenbichler et al., 2022). The same engine that wins for one brand wins for every brand using it.

• Adoption is already mainstream, accelerating the convergence. A Patterns study from Stanford documented that by late 2024, roughly 24% of US corporate press releases were LLM-assisted (Liang et al., 2025). The share of brand communication shaped by the same handful of models is no longer marginal.

• Brand voice itself is converging. Linguistic analysis of marketing copy shows that LLM-signature phrases like “delve into,” “in an era of,” and “navigating the landscape” appear several times more frequently in published marketing text in 2024 than in 2022. Even when individual texts read polished, the collective voice of marketing is becoming statistically recognizable as machine-shaped.

• The cause is mechanical, not cultural. The very technique that makes ChatGPT and Claude usable, reinforcement learning from human feedback, measurably reduces output diversity (Padmakumar & He, 2024; Kirk et al., 2024). The polished corporate voice of generative AI is a statistical mode, and every brand using these models converges toward it.

👉 In a nutshell: When every brand has access to the same models, runs them at marginal cost, and prompts them in similar ways, output differentiation is mathematically excluded. Convergence is no longer a risk. It is the default setting.

The Floor Rises. The Ceiling Does Not.

Generative AI lowers the floor of marketing output dramatically. Anyone with a prompt can produce a usable brief, a decent ad concept, a publishable blog post. Tasks that once required a junior team now take minutes.

The peer-reviewed evidence backs this up precisely. Brynjolfsson, Li and Raymond (2025), in a Quarterly Journal of Economics field study of 5,179 customer support agents, found that AI assistance raised novice performance by 34% while leaving the most skilled performers nearly unchanged. AI is a floor-raising technology by design.

But the ceiling has not moved.

AI models are trained to predict the statistically most likely next token. By construction, they produce the average, not the exceptional. The polished, plausible, well-structured middle of the distribution. That is genuinely useful for most marketing work. It is also exactly why AI cannot produce the brand-defining campaign, the category-creating positioning, or the insight that makes a strategy memorable.

The pattern extends well beyond creative work. The same ceiling effect applies to every strategic discipline in marketing. AI can summarize a category landscape, but it cannot tell you which segment is worth owning. AI can draft a positioning statement that scans well, but it cannot decide whether your brand should fight for differentiation in feature space or in meaning space. AI can list ten innovation directions, but it cannot weigh which one fits your organization’s capabilities, culture, and risk appetite. These are judgment calls that depend on context the model does not have, on consequences the model cannot weigh, and on conviction that comes from years of pattern recognition the model has never lived through.

In a recent conversation with the team at Jung von Matt, one of Germany’s most awarded creative agencies, the point came up exactly this way. Real creative excellence still comes from human experience, taste, and judgment. AI can contribute to every sub-step of the work. It accelerates the brief, generates variations, tests concepts. But the differentiating idea, the one that makes a brand recognizable for a decade, comes from people doing things AI is structurally incapable of doing.

The same is true for strategic decisions. AI can run the analysis. Humans still have to make the call.

👉 In a nutshell: AI raises the floor of competence. It does not lower the cost of excellence. The work that defines a brand, the positioning that wins a category, the strategy that holds for a decade, still requires the rarest resource in marketing: human judgment trained by experience.

The Strongest Counter-Argument

The best case against the homogenization thesis comes from Iansiti and Lakhani at Harvard. In Competing in the Age of AI, they argue that AI-native firms win through superior operating models. Their canonical example: Ant Financial serves over ten times more customers than the largest US banks with less than one-tenth the employees (Iansiti & Lakhani, 2020).

We see the same pattern playing out among startups right now. Companies moving first to build AI-native operating models are unlocking remarkable efficiencies, achieving with small teams what used to require large ones. A handful of engineers and marketers can run operations that would have demanded entire departments three years ago.

That looks like a sustainable advantage. It is not.

Iansiti and Lakhani themselves write that “the software of these new types of company is often open source. Competitive advantage has moved from the production technologies to the data these companies have amassed” (Iansiti & Lakhani, 2020).

Translation: the AI is commodity. The advantage sits in everything around it:

• Data,

• processes,

• organizational design.

That is the five-moat argument in different words. Any advantage built on a technology lasts only until competitors get the same technology. The first movers among AI-native startups are real winners today, but only until competitors recognize the pattern and rebuild around it. AI factories had a meaningful lead five years ago. That lead is compressing now. The moat always migrates to whatever has not yet been commoditized.

Subscribe now

So What Actually Works?

If AI is infrastructure, where does the moat live? The literature converges on five sources:

1. Proprietary data, but only when activated: Brynjolfsson et al.’s (2021) analysis of predictive analytics and Wedel and Kannan’s (2016) Journal of Marketing synthesis show that raw data does not create advantage. Data combined with IT capital, educated workers, and workflow integration generates up to $918,000 in additional sales versus matched competitors. The data is the easy part. The complementary capabilities are the moat.

   Consider Spotify Wrapped: Every streaming service has listening data. Only Spotify turned that data into an annual cultural moment competitors cannot replicate even with identical inputs.

2. Brand and distinctive assets, which matter more as AI mediates discovery: When 80% of consumers trust the brands they use, more than they trust business, media, government, or NGOs (Edelman, 2025), and when LLMs increasingly curate purchase choices, distinctive brand assets become survival infrastructure. Longoni and Cian (2022) found in Journal of Marketing that 76.8% of consumers prefer AI for utilitarian choices but 81.2% prefer humans for hedonic ones. Brand still matters. In some categories, it matters more.

   Patagonia is the cleanest illustration: Decades of consistent environmental activism mean that when a customer asks an LLM for sustainable outdoor brands, Patagonia surfaces first. The brand position is so distinctive that AI amplifies it rather than dilutes it.

3. Trust, which is migrating away from AI vendors and toward brands: Trust in AI companies fell from 61% to 53% globally between 2019 and 2024, and from 50% to 35% in the US (Edelman, 2024). A 13-experiment study in Organizational Behavior and Human Decision Processes found that disclosing AI use systematically erodes trust, including when the AI use is high quality (Schilke & Reimann, 2025). Brands that earned trust before the AI wave have an asset competitors cannot prompt into existence.

4. Human judgment at the jagged frontier: The most-cited workplace AI experiment to date, by Dell’Acqua and colleagues at Harvard, BCG, and MIT (2023), showed that AI-augmented consultants completed 12.2% more tasks, 25.1% faster, and at 40% higher quality on tasks inside AI’s capability frontier. On tasks outside that frontier, they were 19 percentage points more likely to produce wrong answers. The professionals who lost performance “blindly adopt[ed] AI output and interrogate[d] it less.” Judgment about when to trust the machine is the new craft. And the work at the very top, the strategy, the positioning, the brand-defining idea, remains structurally beyond AI, because AI optimizes toward the statistical middle, not the meaningful edge.

5. Speed of organizational learning: An MIT Sloan and BCG study of 3,000+ managers found that only 10% of organizations achieve significant financial benefits from AI, and the differentiator is mutual human-AI learning (Ransbotham et al., 2020). BCG’s 2025 study went further. Only 5% of firms are “future-built,” but they outperform laggards by 1.7× on revenue growth and 3.6× on three-year total shareholder return. They allocate AI investment 10-20-70: 10% to algorithms, 20% to data, 70% to people and processes (Apotheker et al., 2025).

   Future of Marketing is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

The Strategic Test

Here is the question every marketing leader should be asking right now:

If your top three competitors deployed a comparable AI stack to the one you are building right now, similar models, similar tools, similar agents, what would still make your brand the obvious choice for your customers?

If the answer is nothing, you are building infrastructure, not strategy.

If the answer is clear, a unique data asset wired into customer experience, a brand with distinctive assets, trust earned through human accountability, judgment that works at the jagged edge, an organization that learns faster than the team across the street, then AI becomes a multiplier on something defensible.

That is the difference between an AI-enabled brand and an AI-dependent one.

But Not Investing Is Also a Strategic Mistake

A marketing leader might draw the wrong conclusion: if AI is just infrastructure, perhaps the smart move is to wait, let competitors waste their budgets, and pick up the proven tools later. The evidence rules this position out.

Infrastructure decisions, by definition, are not optional. Electricity was not a competitive advantage by 1930, but firms without electricity went out of business. The same logic now applies to AI.

The performance gap between AI leaders and laggards is now measurable and widening. Babina et al. (2024), publishing in the Journal of Financial Economics and drawing on resume data covering 535 million individuals and 180 million job postings, found that a one-standard-deviation increase in firm AI investment is associated with 20.3% higher sales growth, 21.9% higher employment growth, and 22.4% higher market valuation growth. The authors note that “AI-powered growth concentrates among the ex-ante largest firms, leading to higher industry concentration and reinforcing winner-take-most dynamics.” This is not a productivity story. It is a market-structure story.

Customer expectations have shifted irreversibly. Salesforce’s 2024 State of the AI Connected Customer survey of 15,015 consumers and 1,570 business buyers found that the share of customers who feel brands treat them as unique individuals nearly doubled from 39% in 2023 to 73% in 2024 (Salesforce, 2024). The AI-enabled customer experience that was a differentiator 18 months ago is now table stakes.

Talent is migrating toward AI-enabled organizations. Microsoft and LinkedIn’s 2024 Work Trend Index found that 66% of leaders would not hire someone without AI skills, and 71% would prefer a less experienced candidate with AI skills over a more experienced one without (Microsoft & LinkedIn, 2024). The most AI-fluent employees are precisely the ones most willing to leave laggard organizations.

The cost of inaction is not falling behind on AI. It is falling out of the market entirely.

The Implementation Trap: Why Most Investments Fail Anyway

So AI is infrastructure, not advantage. And not investing is fatal. But here is the third side of the triangle that completes the strategic picture: even investing usually fails.

The MIT NANDA initiative’s 2025 GenAI Divide report estimated that of $30 to $40 billion in enterprise GenAI investment, roughly 95% of organizations are seeing zero measurable P&L impact (Challapally et al., 2025). Only about 5% of pilots reach production. McKinsey’s 2025 State of AI survey arrived at a strikingly similar number from a different angle: just 6% of firms qualify as AI high performers (Singla et al., 2025).

For large enterprises, the gap is even more pronounced. This is the territory I explored in The AI Innovator’s Dilemma in Marketing. The pattern Christensen (1997) identified almost three decades ago applies to AI with painful precision. Large organizations have legacy systems, established processes, internal politics, and embedded incentive structures that resist exactly the kind of fundamental rewiring AI requires. They run pilots. They publish announcements. They do not transform.

The diagnosis comes from McKinsey’s own data (Singla et al., 2025): out of 25 organizational attributes tested, workflow redesign has the largest effect on AI EBIT impact. Only 21% of firms have fundamentally redesigned even some workflows. The other 79% are layering AI on top of legacy processes and wondering why the ROI never arrives.

BCG (Apotheker et al., 2025) calls this the 10-20-70 rule: 10% of AI transformation work is the algorithms, 20% is the technology backbone, and 70% is people and processes. Most enterprises invest in the opposite ratio.

The cost of not investing in AI is real and rising. But the cost of investing without organizational transformation is even higher: a fully funded, fully visible failure that drains budget, talent, and credibility.

What This Means for Marketing Leaders

The three findings combine into a single strategic position.

1. AI is infrastructure. It will not, on its own, differentiate your brand. Stop expecting it to.

2. Not investing is not a viable strategy. Laggards face widening performance gaps, customer defection, talent flight, and structural disruption.

3. Investing without organizational rearchitecture is the most expensive mistake of all. The 95% who fail are not failing because they bought the wrong models. They are failing because they bought models and assumed the rest would follow.

The defensible position is to invest in AI as infrastructure so you are not left behind, while investing disproportionately in the things AI cannot commoditize: proprietary data, brand, trust, human judgment, and organizational learning speed. The 10% on algorithms keeps you in the game. The 70% on people and processes is where the moat actually lives.

Subscribe now

Three Moves for the Next 90 Days

If the analysis above is right, the playbook for the next quarter is not a tool list. It is three deliberate moves.

Move 1: Audit your moat, not your stack: Stop counting tools. Map the five sources of defensible advantage against your organization. Where do you have proprietary data that is genuinely activated, not just stored in a CDP? Where is your brand distinctive enough to survive LLM mediation? Where is trust earned rather than claimed? Where do your people exercise judgment AI cannot replicate? Where does your organization learn faster than the team across the street? Each gap is a strategic priority.

Move 2: Redesign one workflow end-to-end before adding another tool: The McKinsey data is unambiguous: workflow redesign has the single largest effect on AI EBIT impact (Singla et al., 2025). Pick the workflow where the gap between current and possible performance is largest. Redesign it from first principles. Then layer AI in. Resist the temptation to deploy three more tools while the existing ones sit half-integrated.

Move 3: Invert your AI investment ratio: Most marketing organizations spend 70% on algorithms and tools, 20% on data, and 10% on people and processes. The 10-20-70 rule of future-built firms inverts that exactly (Apotheker et al., 2025). Pause your next AI tool purchase. Reallocate the budget to training, role redesign, and process work. The visible 30% of the iceberg is where most CMOs invest. The invisible 70% is where the harvest lives.

None of these moves require a new tool. All three require a different theory of where competitive advantage comes from in 2026.

Key Takeaways

1. AI fails Barney’s VRIO test: It is valuable but neither rare nor inimitable. By the framework that defines modern strategy theory, AI cannot be a sustainable competitive advantage (Wingate et al., 2025).

2. The evidence for commoditization is overwhelming: Inference costs collapsed 280× in 18 months and have continued to fall by over 99% across three years, while small models now match what required 142× more parameters two years ago (Maslej et al., 2025, 2026).

3. Marketing is where homogenization hits hardest: AI-assisted content is more similar across brands (Doshi & Hauser, 2024), SEO output is indistinguishable from expert work (Reisenbichler et al., 2022), and 24% of US corporate press releases are already LLM-assisted (Liang et al., 2025).

4. AI raises the floor, not the ceiling: AI improves novice performance by 34% but leaves top performers nearly unchanged (Brynjolfsson, Li, & Raymond, 2025). The work that defines a brand, including positioning and strategic choices, still requires human judgment trained by experience.

5. Five sources of defensible advantage remain: proprietary data activated through complementary capabilities, brand and distinctive assets, trust, human judgment at the jagged frontier, and speed of organizational learning.

6. Not investing is fatal: AI leaders outperform laggards 1.7× on revenue growth and 3.6× on three-year TSR (Apotheker et al., 2025). The cost of inaction is widening, not narrowing.

7. But 95% of AI investments fail to deliver P&L impact: (Challapally et al., 2025). Workflow redesign, not technology selection, is the differentiator. The 10-20-70 rule is the rule that matters: 70% of the work is people and processes.

   ---

If this resonated, the question worth sitting with this week is the strategic test above. Send it to your team. The answers will tell you a lot about where your real moat lives.

Yours,

Prof. Dr. Andreas Fuchs 🦊🎓

Sources

Apotheker, J., de Bellefonds, N., Luther, A., Forth, P., Franke, M., & Kropp, M. (2025, September). The widening AI value gap: Build for the future 2025. Boston Consulting Group.

Babina, T., Fedyk, A., He, A., & Hodson, J. (2024). Artificial intelligence, firm growth, and product innovation. Journal of Financial Economics, 151, Article 103745.

Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99–120.

Brynjolfsson, E., Jin, W., & McElheran, K. (2021). The power of prediction: Predictive analytics, workplace complements, and business performance. Business Economics, 56(4), 217–239.

Brynjolfsson, E., Li, D., & Raymond, L. R. (2025). Generative AI at work. Quarterly Journal of Economics, 140(2), 889–942.

Challapally, A., Pease, C., Raskar, R., & Chari, P. (2025). The GenAI divide: State of AI in business 2025 (Project NANDA Report). MIT Media Lab.

Christensen, C. M. (1997). The innovator’s dilemma: When new technologies cause great firms to fail. Harvard Business School Press.

Dell’Acqua, F., McFowland, E., III, Mollick, E. R., Lifshitz-Assaf, H., Kellogg, K., Rajendran, S., Krayer, L., Candelon, F., & Lakhani, K. R. (2023). Navigating the jagged technological frontier: Field experimental evidence of the effects of AI on knowledge worker productivity and quality (Harvard Business School Working Paper No. 24-013).

Doshi, A. R., & Hauser, O. P. (2024). Generative AI enhances individual creativity but reduces the collective diversity of novel content. Science Advances, 10(28), Article eadn5290.

Edelman. (2024). 2024 Edelman Trust Barometer, Tech sector supplement.

Edelman. (2025). 2025 Edelman Trust Barometer special report: Brand trust, From we to me.

Iansiti, M., & Lakhani, K. R. (2020). Competing in the age of AI: Strategy and leadership when algorithms and networks run the world. Harvard Business Review Press.

Kirk, R., et al. (2024). Understanding the effects of RLHF on LLM generalisation and diversity. In Proceedings of ICLR 2024.

Liang, W., Zhang, Y., Codreanu, M., Wang, J., Cao, H., & Zou, J. (2025). The widespread adoption of large language model-assisted writing across society. Patterns, 6(9), Article 101326.

Longoni, C., & Cian, L. (2022). Artificial intelligence in utilitarian vs. hedonic contexts: The “word-of-machine” effect. Journal of Marketing, 86(1), 91–108.

Maslej, N., Fattorini, L., Perrault, R., et al. (2025). The AI Index 2025 annual report. Institute for Human-Centered AI, Stanford University.

Maslej, N., Fattorini, L., Perrault, R., et al. (2026). The AI Index 2026 annual report. Institute for Human-Centered AI, Stanford University.

Microsoft & LinkedIn. (2024). 2024 work trend index annual report. Microsoft Corporation.

Padmakumar, V., & He, H. (2024). Does writing with language models reduce content diversity? In Proceedings of ICLR 2024.

Ransbotham, S., Khodabandeh, S., Kiron, D., Candelon, F., Chu, M., & LaFountain, B. (2020). Expanding AI’s impact with organizational learning. MIT Sloan Management Review and Boston Consulting Group.

Reisenbichler, M., Reutterer, T., Schweidel, D. A., & Dan, D. (2022). Frontiers: Supporting content marketing with natural language generation. Marketing Science, 41(3), 441–452.

Salesforce. (2024). State of the AI connected customer (7th ed.). Salesforce.

Schilke, O., & Reimann, M. (2025). The transparency dilemma: How AI disclosure erodes trust. Organizational Behavior and Human Decision Processes, 188, Article 104405.

Singla, A., Sukharevsky, A., & Yee, L. (2025). The state of AI in 2025: Agents, innovation, and transformation. McKinsey & Company.

Wedel, M., & Kannan, P. K. (2016). Marketing analytics for data-rich environments. Journal of Marketing, 80(6), 97–121.

Wingate, D., Burns, B. L., & Barney, J. B. (2025, May 8). Why AI will not provide sustainable competitive advantage. MIT Sloan Management Review.

• It is written for marketing leaders, not for designers or developers. Every example is a real marketing problem: a launch landing page, an customer pitch deck, a campaign video, a trade show booth.

• Brand is the centerpiece, not an afterthought. Section 4 gives you six concrete ways to plug your existing brand into Claude Design, from a simple URL to a full design system extracted with Cowork.

• Four real-world use cases, not hypothetical demos. Two from my own work, two from rhome, a Global Mobility startup. With the exact prompts and screenshots from each step.

• Honest limitations included. Three weeks of testing surfaced real rough edges. They are in here, not buried.

The first half is free. The full brand framework, the four use cases, the iteration system, and the limitations are for premium subscribers.

Subscribe now

Part 1: 20 Ways to Use AI as Your Marketing Co-Worker: How to set up Claude Cowork, how to work with it, and 20 use cases to start today.

Part 2: From Data to Decision in Minutes: How Claude Charts changes the game for marketing and sales.

Part 3: Skills are the New Skills in Marketing: How custom Skills turn Claude into your personal marketing engine.

Part 4 (this guide). Claude Design for Marketing: How marketing teams build websites, decks, and campaigns by talking to Claude.

A month ago, I showed you how custom Skills make Claude permanently better at your specific marketing work.

This guide picks up where that one left off. Because last week, Anthropic announced another product that quietly changes the economics of marketing work. Claude Design. A direct shot at Figma, Canva, and PowerPoint.

The speed at which Anthropic is shipping is breathtaking. Code. Cowork. Charts. Skills. Agents. And now Design. Each one of these would be a standalone company anywhere else. Here, they are connected features of one growing system.

For marketing teams, this is the most important launch since Claude Cowork.

Not because “AI can now design” (that story is old), but because of what Claude Design actually does in the hands of a non-designer:

• Describe a campaign landing page in words. Get a working high-fidelity prototype in under two minutes.

• Drop a website URL, a Figma export, or a folder of past campaigns. Claude reads them and builds in your visual language.

• Paste an article. Get an animated explainer video. Then convert it into a slide deck.

• Point at any world-class brand (Stripe, Linear, Mastercard) and design with their visual standards as a reference.

• Export to Canva, PowerPoint, PDF, standalone HTML, or hand off directly to Claude Code.

One conversation. One brand. One export menu that connects to the tools your team already uses.

Here is what changes for marketing teams:

Your people stop briefing. They start building. The product marketer who needs a campaign landing page does not write a design brief and waits two weeks. She opens Claude Design, points it at your brand, types a prompt, and ships a high-fidelity prototype in fifteen minutes. The same is true for the demand generation lead who needs A/B variants for a hero page, the founder who needs an customer sales deck on a plane, the field marketer who needs a coordinated trade show booth, and the CMO who needs a board deck by Friday morning.

Four properties make this possible:

• Self-service: No designer queue. No external agency. No three-round feedback loop before a draft exists. The person who needs the asset is the person who builds it.

• 100% on-brand: With one of the brand paths in Section 4, every output applies your typography, colors, tone, and layout conventions automatically. You do not get a design that looks like Claude. You get a design that looks like yours.

• Minutes, not weeks: A campaign page in fifteen minutes. A pitch deck in twenty. A trade show booth set in thirty. The bottleneck is no longer execution. It is knowing what to build.

• Marginal cost: No new designer hire. No new tool subscription. No new project budget. The team you have, the brand you have, the Claude plan you already pay for.

This is not a tool that replaces designers. Designers will keep doing what they do better than any model: taste, judgment, brand stewardship, the ability to break a rule for the right reason. They move from execution to direction. They become curators and quality gates, not the bottleneck. For premium brand assets, for category-defining work, for the moments where every pixel matters, you still want a designer in the loop.

But for the 80 percent of marketing assets that today get stuck in queues, eat agency budgets, and arrive too late to matter? Your team can ship them this afternoon.

I have been testing Claude Design intensively since launch. This guide shows you how to set it up, the principles that separate brand-fit results from generic ones, four real field-tested use cases (two of them from my own work), and where it still falls short.

A short note before we start. The tests in this guide take about half an hour to run yourself. Reading about Claude Design gives you half the value. Actually using it gives you the other half. Block the time.

1. How to set up Claude Design

Claude Design is not a new tab inside the Claude chat. It is a separate product with its own URL, its own interface, and its own workspace.

You find it at claude.ai/design. It runs on Claude Opus 4.7, the strongest vision model Anthropic has shipped, and it is currently labeled as a research preview.

• If you are on Pro or Max plan: Go to claude.ai/design and sign in with your existing Claude account. The free tier does not include Design.

  

• If you are on Team or Enterprise plan: Your workspace admin needs to enable it. According to Anthropic’s official documentation, the path is: Organization settings → Capabilities → Anthropic Labs → toggle Claude Design on.

  If Claude Design is not visible after the toggle is on, the rollout is gradual. Try again in a few days. If you are not an admin, this is the IT request you send today.

Two things to know before your first prompt

First, this consumes tokens fast. A single complex landing page can burn through a meaningful chunk of your monthly Pro allowance. If you plan heavy use, Max is the more realistic plan. Team and Enterprise admins should expect higher token consumption per active user than with the standard Claude chat.

Second, treat the research preview label as honest. Some features are still rough. The “Send to Canva” export worked unreliably in my tests. Some exports lose minor formatting. The product is improving weekly, but plan for it.

Subscribe now

2. The interface in three minutes

Open claude.ai/design. You see four entry points at the top of the workspace:

• Prototype for landing pages, websites, web apps, dashboards

• Slide deck for presentations of any kind

• From template for animated videos, design systems, recurring formats

• Other for everything in between (campaign assets, posters, ad creatives, social content)

For each entry point, you write a prompt, attach optional context (brand assets, screenshots, Figma exports, markdown files, URLs), and Claude generates a draft. From there, you iterate. Structural changes happen in the chat. Pixel-level changes happen on the canvas. Variant generation happens through the Tweaks button.

The export menu (top right) gives you six output formats: Canva, PPTX, PDF, standalone HTML, a handoff to Claude Code for development, and a download as ZIP.

That is the whole product, conceptually. The skill is in how you use it.

3. Three principles for getting brand-fit results

This is the part most coverage skips. And it is the reason most AI design output looks the same: clean, modern, competent, and entirely interchangeable.

Based on my testing outcomes, three principles separated good results from generic ones.

Principle 1: Always start from something, not from nothing

Claude Design is dramatically easier to work with when you start from existing material than from a blank slate.

If you have a website, a Figma file, a brand guidelines PDF, a folder of past campaign assets, or even just two or three screenshots of designs you like, Claude Design has a reference. It can match. It can extend. It produces work that looks like yours.

If you start from absolutely nothing, the experience changes. Claude does its best, walks you through smart questions (see screenshot below for landing page example) about audience and tone, and you can hand the design decisions over entirely if you want. The output is competent. But hitting exactly what you imagined takes more iteration, more re-prompting, and more taste calls from you.

This matters for one reason: most marketing teams already have brand assets. Logos, recent campaigns, an existing site, recurring colors, a tone of voice that lives in old emails and decks. The shortcut is not to invent a brand for Claude. It is to extract the one you already have. The next section shows you exactly how, with six concrete paths.

Principle 2: Specificity beats sophistication

A vague prompt produces a generic page. A precise prompt produces a sharp one. The skill is not in finding clever prompts. It is in knowing what to specify.

Four questions are worth answering in every prompt:

1. Who is this for? (Audience and intent)

2. What does it need to do? (Goal: convert, inform, convince, inspire)

3. What should it look and feel like? (Tone, visual reference, examples)

4. What is non-negotiable? (Brand colors, mandatory components, structural constraints)

When all four are present, the first generation is usually 80 percent there. When any of them is missing, you spend three rounds of iteration filling in what you should have specified upfront.

Principle 3: Iterate inside one session, not across new sessions

This one is counterintuitive. When a generated design feels off, the instinct is to start over with a better prompt. Resist that. Stay in the same chat.

Claude Design carries context within a session. It remembers your earlier choices, your brand context, your tone preferences, the structural decisions it has already committed to. When you say “make the hero more confident” or “show me three alternative pricing layouts,” Claude moves forward inside the world it has already built. When you start a new chat, you are starting from scratch.

The faster you converge, the less you spend on tokens, and the closer the output stays to what you actually wanted.

Subscribe now

4. Six ways to give Claude Design your brand

👉 Premium subscribers unlock: The complete six-path framework for plugging your brand into Claude Design (URL, Figma, screenshots, asset folders, borrowed brands, from scratch). Four real field-tested use cases with the exact prompts and screenshots from every step. The four iteration moves that turn a draft into a shippable asset. The export decision tree. The honest limitations. And what this all means for the structure of the marketing function in 2026.

Read more

Why read this article: It explains why 88% of enterprises adopt AI but only 5-6% generate real business value from it, and why the root cause is organizational, not technological. The article applies Clayton Christensen’s Innovator’s Dilemma framework to the current AI adoption crisis in marketing, identifies five structural barriers that reinforce each other, and provides a research-backed playbook for marketing leaders who want to escape the trap before the gap to AI-native competitors becomes irreversible.

Over the past weeks, I've written a lot about what AI can do for marketing. The Claude for Marketing Mini Series or practical guides. And every single time I talk to someone working inside a large organization about these topics, I hear the same thing: "That's great, but at our company, none of this works."

A friend of mine works at Mercedes-Benz. His company rolled out Microsoft Copilot as their official AI tool (like most large enterprises with Microsoft 365, it was the path of least resistance). Deployed, available, supported by IT.

He says it’s useless.

Not in a hyperbolic way. In a very literal way. The tool sits on top of the organization like a disconnected shell:

• It can’t access internal tools or databases.

• It can’t read strategy documents.

• It doesn’t know what projects are running.

• It has no idea what the company decided last quarter.

Now here’s the part that should worry every marketing leader reading this.

If a five-person startup decided today to deploy AI across their entire operation, connect it to every system and data source they have, and start working with it as an integrated part of their team, they could realistically do it by tomorrow afternoon.

Mercedes can’t do that. Neither can BMW, Siemens, Unilever, or most Fortune 500 companies. Not because they lack budget (enterprises spent $37 billion on generative AI in 2025 alone). Not because they lack talent. And not because the technology isn’t ready.

They can’t do it because their own organizations won’t let them.

That’s the trap. And Clayton Christensen diagnosed it almost 30 years ago.

The Framework: Christensen’s Innovator’s Dilemma

In 1997, Harvard professor Clayton M. Christensen published The Innovator’s Dilemma (1997). His core argument remains one of the most powerful ideas in strategy: good management itself causes failure when companies face disruptive technologies. His RPV framework explains why: Resources (people, capital, tools) are flexible, you can redirect them. But processes and values (how decisions get made, what gets prioritized) become embedded in organizational culture. They resist change even when the people inside the organization want to change. The very practices that drive success in established markets are the ones that reject disruptive technologies. AI in the enterprise is following this pattern with remarkable precision.

Henderson and Clark (1990) added a critical dimension: firms fail not because they can’t build or buy new components (the AI tools exist), but because they can’t reconfigure the relationships between components. AI changes how data, creative, targeting, measurement, and optimization relate to each other. That makes it deceptively dangerous for incumbents, because it looks incremental on the surface.

The Paradox: Everyone Adopts AI, Almost No One Benefits

The data on enterprise AI adoption tells a paradox that Christensen would immediately recognize.

88% of organizations now use AI in at least one business function, and Gen AI adoption has reached 72%. But only 39% report any EBIT impact at the enterprise level, and most of those report less than 5%. Only about 6% qualify as “AI high performers” (McKinsey, 2025).

The picture gets worse the closer you look. While 75% of executives rank AI among their top three priorities, only 5% of companies achieve AI value at scale (what BCG calls “future-built”). A full 60% are laggards reporting minimal gains despite substantial investment. Less than 30% of CEOs were satisfied with their AI returns (BCG, 2025).

Generative AI has entered the Trough of Disillusionment on Gartner’s 2025 Hype Cycle, with over 30% of GenAI projects abandoned after proof of concept. And across the enterprise landscape, 95% of GenAI pilots fail to deliver measurable P&L impact, despite $30-40 billion in total investment (MIT NANDA, 2025).

Marketing sits at the epicenter of this paradox. It has been the #1 function for AI adoption for eight consecutive years (McKinsey, 2025). More than half of all enterprise GenAI budgets flow into marketing and sales tools (Menlo Ventures, 2025). Yet most of that spending produces no measurable financial return.

👉 Adoption was never the issue. What’s missing is impact.

Why Enterprises Can’t Use AI: Five Barriers That Reinforce Each Other

The barriers preventing large enterprises from using AI’s full potential in marketing aren’t independent obstacles. They form a self-reinforcing system where each barrier makes the others harder to overcome.

Barrier 1: The Security Wall

The single largest barrier is fear of data leakage. And the fear is well-founded.

34.8% of corporate data entered into AI tools is sensitive, up from 10.7% two years earlier (Cyberhaven, 2025). GenAI now accounts for 32% of all corporate data exfiltration, making it the #1 vector for data leaving enterprise controls (LayerX, 2025).

The Samsung incident crystallized these fears into policy. In March 2023, Samsung’s semiconductor division allowed employees to use ChatGPT. Within 20 days, three separate leaks occurred: an engineer pasted proprietary source code, another submitted equipment test sequences, and a third uploaded meeting transcripts. Samsung banned all generative AI tools company-wide. JPMorgan Chase, Goldman Sachs, Apple, Amazon, and Verizon followed within months.

Today, 56% of enterprises block most AI tools while maintaining narrow allowlists. Enterprise procurement cycles stretch 3-12 months for each new tool. By the time something clears security audits, SOC 2 validation, and legal review, it may already be obsolete. AI capabilities launch weekly. Approval takes quarters.

Barrier 2: The Data Architecture Trap

Even when security clears the path, enterprise marketing data is almost never in a state that AI can use.

Only 7% of enterprises consider their data “completely ready” for AI (Cloudera & Harvard Business Review, 2026). The average enterprise marketing organization runs 91-120 distinct martech tools, selected from a landscape of 15,384 commercial solutions (Brinker, 2025). And the average enterprise uses 897 applications, with only 29% integrated (MuleSoft, 2025). That means 71% of enterprise applications exist in silos, invisible to AI.

Data preparation consumes roughly 60% of a typical AI project’s timeline. And as Gartner warns: “High-quality data does not equate to AI-ready data.” AI-ready data requires lineage tracking, diversity, low-latency delivery, and discoverability across systems. Most enterprise marketing data meets none of these requirements.

Meanwhile, a startup on a modern cloud-native stack connects all data sources to AI on day one. No silos to cross, no integration projects to fund.

Subscribe now

Barrier 3: Organizational Quicksand

The most underappreciated barriers are human, not technical.

By 2026, 20% of organizations will use AI to flatten structures, eliminating more than half of current middle management positions (Gartner, 2025). This creates rational resistance. Marketing middle managers who oversee campaign setup, A/B testing, performance reporting, and content workflows see their core functions being automated. Gartner observed an “AI Dropout Effect”: employees who perceive AI as a threat to their identity disengage, burn out, or leave.

The skills gap compounds the problem. Only 2.6% of marketers identify as AI experts (CoSchedule, 2026). And nobody owns the outcome: the CFO expects headcount reduction, the CMO wants capacity expansion, the CEO wants growth. Without unified ownership, AI pilots succeed technically but fail organizationally.

Barrier 4: Hidden Costs

Enterprise AI costs dwarf their sticker prices. Total cost of ownership runs 3-5x the advertised subscription price when accounting for integration, customization, training, and maintenance. For every $1 of tangible technology investment, companies spend up to $10 on intangible costs like process redesign, reskilling, and organizational transformation (Pereira et al., 2026). And 77% of the hardest challenges in AI deployments turn out to be these intangible costs, not the technology itself.

Barrier 5: Legacy Systems and Technical Debt

68% of IT decision-makers say legacy systems prevent their organization from fully embracing AI (Pegasystems, 2025). Technical debt constrains AI success for 81% of executives, consuming up to 29% of AI budgets (IBM, 2025). An estimated 60-80% of enterprise IT budgets go to maintaining existing systems, leaving little for innovation.

The Inevitable Result: Shadow AI

When official AI tools are restricted, slow, or disconnected from useful data, employees route around them.

73.8% of workplace ChatGPT usage occurs on personal accounts lacking enterprise privacy controls (LayerX, 2025). Shadow AI is highest in marketing and sales teams (UpGuard, 2024). And shadow-AI-associated data breaches cost organizations $670,000 more on average (IBM, 2025).

The organization creates the restriction.

The restriction creates the workaround.

The workaround creates the risk.

That’s the cycle.

And nowhere is this cycle more visible than in the tools enterprises have already deployed.

Subscribe now

The Dumb Copilot Problem

The experience at Mercedes is symptomatic. The data shows it’s happening everywhere.

Microsoft 365 Copilot, priced at $30/user/month, has achieved only 3.3% penetration across Microsoft’s ~450 million commercial M365 subscribers. Its paid subscriber market share fell from 18.8% to 11.5% between July 2025 and January 2026, while ChatGPT grew to 55.2%.

Even Microsoft CEO Satya Nadella is frustrated. In December 2025, he emailed engineers stating that Copilot’s connections with services like Gmail and Outlook “for the most part don’t really work.” Perhaps most telling: approximately 75% of employees in Microsoft’s own Copilot division paid out of pocket for ChatGPT because Microsoft wouldn’t expense the rival subscription. Salesforce’s Agentforce tells a similar story: only 5,000 deals after launch, pricing changed twice within a year.

The tools have potential. But potential locked inside organizational constraints produces nothing.

Meanwhile, Startups Deploy AI in Days

The speed differential between startups and enterprises goes well beyond incremental. We’re looking at an order of magnitude.

Enterprise AI projects average 17 months from initiation to production (McKinsey, 2025). Startups deploying focused AI use cases achieve working results in 1-3 weeks. And the market consequences are already visible: AI-native startups now capture 63% of the AI application layer, up from 36% the prior year, earning nearly $2 for every $1 incumbents earn (Menlo Ventures, 2025).

The Cursor vs. GitHub Copilot case captures the dynamic perfectly. Cursor, an AI-native code editor built by four MIT graduates with roughly 150 employees, reached $2 billion ARR by Q1 2026, competing directly with GitHub Copilot, backed by Microsoft’s $37.5 billion AI investment. The company reached $200M before hiring a single enterprise sales representative.

The efficiency numbers are staggering. AI-native startups average $3.48 million in revenue per employee, approximately 5.7x higher than the $611K average at leading traditional SaaS firms. AI companies grow 4x faster while operating with 7-8x fewer employees per dollar of revenue (Emergence Capital, 2025).

And here’s the most counterintuitive finding. OECD data from 2025 shows that for marketing and sales purposes specifically, small enterprises using AI exhibit a slightly higher utilization share than large ones. It’s the first technology adoption cycle in history where small firms don’t trail large ones in the most commercially important function.

That’s the Innovator’s Dilemma, measured in real time.

A Playbook for Marketing Leaders (What the 5% of Companies Do Differently):

The gap between the 5% creating value and the 60% stagnating was never about the AI model. It was about the organization. And the evidence on what works is remarkably consistent across sources. Here are five guiding principles:

Premium subscribers unlock: The complete 5-principle playbook based on research from McKinsey, BCG, and the Stanford Enterprise AI Playbook. Including the single strongest predictor of AI impact (Principle 1), BCG's 10/20/70 investment rule with a visual breakdown, a concrete lighthouse project framework for quick wins in 2-4 weeks, Christensen's autonomous unit model applied to marketing, and a minimum viable data foundation approach using MCP.

Read more

This is a practical guide to building a AI-powered CMO Cockpit: a five-layer decision system that replaces dashboards, spreadsheets, and Monday morning guesswork. You will see exactly what goes into each layer (forecasts, playbook, KPIs, operations, market intelligence), how to connect real data sources via MCP (Notion, Slack, CRM, and more), and how AI turns raw data into recommendations. The article includes an interactive prototype you can click through yourself: every tab, every filter, every forecast is live. Inspired by a conversation with a tech startup CEO, I built the whole thing with Claude in a single afternoon.

Stop Staring at Dashboards. Start Steering.

Let me skip the part where I tell you dashboards are important. You already have one. Probably several.

The question worth asking is: when was the last time a dashboard actually changed a decision you made?

Most CMOs I talk to describe the same experience. The numbers are there. Revenue, leads, pipeline, cost per acquisition. Neatly formatted. Updated weekly. And completely disconnected from the question that matters on Monday morning: what should we do differently?

That disconnect has a price. In a recent survey of 678 executives, 63% of CMOs said they miss opportunities because they cannot make decisions fast enough (PwC, 2025). Not because they lack data. Because the data lives in one place and the decisions happen in another.

Meanwhile, 84% of marketing organizations remain stuck in what IBM calls “pilot purgatory” when it comes to AI. Only 19% of pilots have delivered expected ROI (IBM IBV, 2025). The tools are getting smarter. The organizations are not keeping up.

Dashboards were built for an era when the primary challenge was visibility. That era is over. What marketing leaders need now is a system that connects numbers to actions, actions to owners, and owners to outcomes.

That system is a cockpit.

What a CMO Cockpit Actually Is

A cockpit is what happens when you stop treating data as something to look at and start treating it as something to steer with.

Think about how a pilot works. There is no screen in one room showing altitude and a separate room where you adjust the engine. Everything is in one place: current state, trajectory, instruments, controls. You see where you are, where you are heading, and you can intervene immediately.

A CMO Cockpit follows the same logic. It brings five layers together into one decision environment:

Layer 1: Headquarters. The morning briefing. Where are we? What needs attention? Where are we heading? A multi-KPI forecast that shows year-end projections across three scenarios (optimistic, base, conservative) for revenue, leads, pipeline, win rate, and CAC. Alerts that combine KPI anomalies with signals from inside the organization. Quarterly milestones that show what has been achieved and what is at stake. Everything a CMO needs to orient, on one screen.

Layer 2: Playbook. The bridge between analysis and action. What should we do this week, this month, this quarter? Each action prioritized, tied to a goal, assigned to an owner. Fed by an interactive quarterly review (wins, misses, decisions) that the system generates and the CMO can edit. Below that, goal coverage: for each target, are the planned measures sufficient? If projected leads fall short even with all measures active, the cockpit flags it now, not in the quarterly review.

Layer 3: Performance. KPIs and budget in one view, because they belong together. Sixteen metrics across pipeline, leads, efficiency, and engagement. Revenue versus prior year. Channel performance with CPL, conversion rates, and trend sparklines. An interactive budget allocator that lets you model: what happens if we shift 30% of the event budget to content? The system recalculates projected leads in real time.

Layer 4: Operations. The layer most dashboards ignore entirely. Every running campaign, event, and program with the KPIs that actually matter for that specific activity. LinkedIn Ads shows impressions, CTR, CPL, conversion rate. An event shows meetings booked, leads captured, pipeline generated. Outbound shows emails sent, reply rate, SQLs. Filterable by type, status, owner. Plus project tracking, team workload with capacity indicators, AI agent reliability monitoring, and a marketing calendar with preparation status.

Layer 5: Market Insights. The outward-looking layer. Competitive intelligence with funding rounds, product moves, and (crucially) which customers they won and which they lost to you. Market signals with strategic implications. Organization signals aggregated from Slack, Teams, Notion: product shipped a feature, CEO confirmed pricing changes, a customer escalated. Pipeline intelligence across four quadrants: hot, at-risk, churn watch, expansion.

These five layers form a loop. A forecast that shows a lead gap triggers a playbook action. That action needs budget. The budget needs someone (or an agent) to execute. And market signals might reshape the whole strategy before the quarter is over. Every layer feeds the next. Cut one and the system breaks.

This is what moves CMOs from monitoring to steering.

👉 Premium subscribers unlock: How to actually build this cockpit, step by step. The technology shift that makes it possible now (Cowork, MCP connectors, Skills). A real build story showing what Claude produced in a single afternoon. The three evolutionary stages from cockpit to Virtual CMO. A six-step roadmap with practical guidance on connecting data sources. The full interactive prototype to click through. Plus the Key Takeaways.

Upgrade to Unlock the Full Guide

Read more

Part 1: 20 Ways to Use AI as Your Marketing Co-Worker: How to set up Claude Cowork, how to work with it, and 20 use cases to start today.

Part 2: From Data to Decision in Minutes: How Claude Charts turns marketing data into interactive visualizations and dashboards.

Part 3 (this guide) Skills are the New Skills in Marketing: How to create custom Skills that make Claude permanently better at your specific marketing work … with 10 ready-to-use Skill templates.

Subscribe now

Our whole life has been shaped by one thing: building skills.

We went to school. We went to university. We did internships, trainee programs, first jobs where we had no idea what we were doing. We watched people who were better than us and tried to figure out what they knew that we didn’t. We read books. We took courses. We failed, got feedback, and tried again. And slowly, over years, we got good at things.

Maybe you were the one on the team who built the best presentations. The person everyone came to when a pitch had to land. Maybe you were the analyst who could pull insights out of messy data that nobody else even looked at. Maybe you were the brand strategist who could articulate a positioning in one sentence while the rest of the room was still debating slide three.

Some of us became the go-to person for market research. The one who knew how to design a survey that surfaces real customer insights instead of confirming what everyone already believed. Others became the best copywriter in the department, the one who could write a headline that stops the scroll and a CTA that converts. Others got really good at briefing agencies, reading a P&L, running focus groups without leading the answers, or segmenting audiences in ways that made campaigns perform twice as well.

These skills were our edge. The thing that separated us from the next candidate, the next freelancer, the next agency pitch. They made us valuable. They made us hard to replace.

And they didn’t come for free. Every one of them took years.

The good news: learning never stops. The marketer who’s still curious, still willing to sit with something unfamiliar until it clicks, has always had an advantage. That hasn’t changed.

Subscribe now

What has changed is what we need to learn next.

Technology is reshaping what marketing teams look like, how they’re staffed, what gets outsourced, and what gets automated. Organizations are restructuring. Budgets are tighter. Teams are smaller. “Do more with less” stopped being a management cliché and became the operating reality for most marketing departments.

And right in the middle of this: AI.

We’ve created something that can execute many of the tasks we spent years mastering. Not because it’s smarter than us. But because it doesn’t forget, doesn’t get tired, doesn’t have an off day. It applies the same quality standard to the 500th campaign brief the way it did to the first.

The marketer who spent years mastering Excel? Claude builds the spreadsheet from a description. The one who could write a positioning statement in her sleep? Claude drafts five variants before she finishes her coffee. The one who was fastest at turning a report into a slide deck? Claude does it in two minutes.

This can feel unsettling. But here’s how I see it:

The competition hasn’t stopped. It has shifted. The people who will stand out in marketing won’t be the ones who resist the tools. They’ll be the ones who master them faster than everyone else. Who figure out how to take what they know (all those skills, all that experience, all that judgment built over years) and encode it into systems that scale.

And this isn’t just about individuals.

Think about marketing organizations. Every team has built its own way of doing things over the years. How campaigns get briefed. How reports get structured. How brand guidelines get applied. How content gets repurposed across channels. Most of this knowledge lives in people’s heads, in scattered documents nobody maintains, or in the muscle memory of whoever has been around the longest.

When that person leaves, the knowledge walks out with them. When a new team member joins, it takes months before they produce work at the level the team expects. Every organization knows this problem. Very few have solved it.

The same shift that applies to individuals applies to teams: the marketing organizations that will pull ahead aren’t the ones with the biggest budgets or the most people. They’re the ones that figure out how to capture what makes them good and turn it into infrastructure that works independently of who’s in the room on any given Monday.

The race to be the best hasn’t ended. The track has changed. For you personally, and for your team.

And the most valuable skill in the next chapter of marketing might be: building Skills.

Not metaphorically. Literally.

That’s what this guide is about.

Subscribe now

What Are Skills? (And Why Should Marketers Care?)

A Skill is a set of instructions you write once that Claude follows from that point forward.

Unlike a prompt that disappears when the chat ends, a Skill lives permanently in your Claude setup. It gets loaded automatically whenever Claude recognizes that your request matches what the Skill was built for.

Think of it this way: a prompt is a conversation. A Skill is a playbook.

If you’ve spent the last two years getting better at prompting, here’s the uncomfortable truth: that effort was necessary, but it doesn’t compound. Every prompt you’ve ever written is gone the moment you close the chat. Skills change that equation. You invest the thinking once. You define what good looks like once. And from that point forward, Claude delivers at that level without you having to explain it again. The game isn’t writing better prompts anymore. It’s writing better Skills.

You don’t tell your senior content strategist how to write a LinkedIn post every time you ask for one. They know your voice, your format, your audience, your no-go phrases. They just deliver. That’s what a Skill does for Claude.

Anthropic provides pre-built Skills for common tasks (creating PowerPoint files, Excel spreadsheets, Word documents, PDFs). Those are useful. But the real power is in custom Skills: the ones you build yourself, tailored to your exact marketing workflows.

Here’s why this matters for marketing specifically:

Marketing work is highly repetitive in structure but highly variable in content. Every campaign brief follows the same logic but covers different products. Every content calendar uses the same framework but different topics. Every competitive analysis follows the same structure but different markets.

Skills capture the structure. You provide the content. Claude handles the rest.

👉 Skills are the difference between using AI as a chatbot and using AI as a trained team member.

Subscribe now

How Skills Work (The Technical Part, Made Simple)

A Skill is just a folder with a text file inside.

That file is called SKILL.md. It contains two things:

1. Metadata (a name and a description that tells Claude when to use this Skill)

2. Instructions (the detailed playbook Claude follows when the Skill is activated)

Here’s the simplest possible example:

---
name: campaign-brief
description: Create structured marketing campaign briefs. Use when the user asks for a campaign brief, campaign plan, or marketing brief.
---

# Campaign Brief Generator

## Instructions
When creating a campaign brief, always include these sections:
1. Campaign objective (one sentence, measurable)
2. Target audience (specific segment, not "everyone")
3. Key message (what we want them to think, feel, do)
4. Channels and tactics
5. Timeline and milestones
6. Budget allocation
7. KPIs and measurement plan

## Rules
- Keep the objective to one sentence
- Always ask for the product/service if not provided
- Use plain language, no marketing jargon
- Include a "What success looks like" section at the end

That’s it. This file teaches Claude how to create a campaign brief every single time, without you having to explain the format again.

The description field is critical. Claude reads every Skill description to decide which ones are relevant to your current request. If you ask for a campaign brief, Claude sees the description, loads the Skill, and follows the instructions. Automatically.

You can also invoke a Skill directly by typing /campaign-brief in Claude Cowork.

Subscribe now

How to Set Up Your First Marketing Skill

Every article takes days to research and write. Most of them are free. If they help you stay ahead in marketing, consider becoming a paid subscriber to support this work.

👉 Premium subscribers unlock: Three ways to set up your first Skill (including a prompt that lets Claude build it from your own best work). Plus 10 complete marketing Skill templates with full SKILL.md code to copy and install: Brand Voice, Content Repurposing, Competitive Intel, Email Sequences, Campaign Post-Mortems, SEO/GEO, Customer Personas, Budget Allocation, Leadership Reporting, and Sales Enablement. Plus a framework for turning Skills into organizational infrastructure and a concrete action plan for your first week.

Read more

Part 1: 20 Ways to Use AI as Your Marketing Co-Worker: How to set up Claude Cowork, how to work with it, and 20 use cases to start today.

Part 2 (this guide): From Data to Decision in Minutes: How Claude Charts turns marketing data into interactive visualizations and dashboards.

Part 3: Skills are the New Skills in Marketing: How to create custom Skills that make Claude permanently better at your specific marketing work … with 10 ready-to-use Skill templates.

Two weeks ago, I showed you how to use Claude Cowork as your marketing co-worker. That guide was about documents, briefs, and workflows. It became the most-read article on this channel.

This one picks up where it left off.

Be honest. How many hours did your team spend last month turning data into slides?

Not analyzing data. Not making decisions. Just formatting. Copying numbers into Excel. Screenshotting dashboards. Rebuilding the same bar graph for the third stakeholder who wanted a “slightly different view.”

Nobody planned for this to become a core marketing activity. But here we are. The gap between having an insight and showing it to someone who needs to act on it has quietly become one of the biggest time sinks in modern marketing.

Now imagine the opposite. You have an idea for a chart. You describe it. And it exists. Interactive, clean, ready to share. In minutes. You spend your time on what the data means, not on how to display it. Your leadership team sees exactly what you see. Your decisions get faster because the visual is no longer the bottleneck.

That is what Claude Charts does.

Click to View Interactive Dashboard

How It Works

Open claude.ai. Type a prompt that starts with something like “Create an interactive chart about …” and describe what you need.

That is it.

Claude generates the visualization directly in the chat. You can hover over data points, explore segments, and interact with the chart in real time.

One important detail: if Claude tries to write code instead of generating a chart, add “Don’t code. Interactive chart only.” to your prompt. Works every time.

Share

Claude Charts Goes Far Beyond Dashboards

When most people hear “interactive charts,” they think bar graphs and pie charts. That is only the beginning.

Claude builds full interactive visualizations. Conceptual models. Frameworks with clickable scenarios. Teaching tools you can use in a workshop or a lecture.

Here is an example: I needed an interactive teaching module for my university lecture on marketing communication. The topic: how does a brand message change when it passes through a media multiplier? One-step vs. two-step communication.

One prompt. Claude built an interactive HTML application with four real-world scenarios (TV advertising, influencer marketing, PR, retail), animated message flows, a message fidelity meter that shows control loss at each stage, and discussion questions. Ready to use in the browser, in front of 200 students.

Click to View Interactive Visualization

This is not a chart. This is an interactive application. Built in a single conversation.

Here is another example: I wanted a visual overview of every essay I have published on Future of Marketing. Not a spreadsheet. Not a Notion table. An interactive timeline.

Claude built a full HTML application: 16 essays mapped chronologically from October 2025 to March 2026, grouped by month, with a vertical timeline. Four statistics tiles at the top (total essays, active months, publishing frequency, peak month). Six filterable theme tags (AI & Agents, Strategy, MarTech & Data, Org Design, Content & Growth, Newsletter Meta). Click a tag, everything else dims. Click again, everything returns. Every entry links directly to the original essay. Works in light and dark mode.

Click to View Interactive Timeline

One prompt. A content library you can actually navigate.

💡 Pro tip: You do not need to type in your articles manually. Use the Claude in Chrome extension, give it your website URL, and Claude scrapes the page and extracts every article automatically. In my case: futureofmarketing.de. That is it. Alternatively, connect a Notion database, an Excel file, or any other source where you already track your content. Claude reads the database and builds the timeline from there. Either way: one input, zero manual data entry.

For marketers, the implications are immediate. Client presentations where stakeholders can click through scenarios. Workshop materials where participants explore frameworks themselves. Training modules for new team members. Content archives your audience can browse. Strategy visualizations that do not just show data, but show how things work.

👉 Think beyond dashboards. Every framework, every model, every process you explain in meetings can become an interactive artifact.

The Real Power: Connect Your Actual Data

The examples below use manually entered numbers. That is fine for a quick visualization. But the real leverage starts when you connect Claude to the tools where your data already lives.

Through Connectors and Plugins, Claude can pull data directly from your marketing stack. No exporting. No copy-pasting. No reformatting CSVs at 11pm before a Monday meeting.

Here is what you can connect today:

• Productivity & Knowledge Management: Google Drive, Google Sheets, Notion, Microsoft Excel, Confluence. Upload a spreadsheet or connect a Notion database, and Claude charts your data directly.

• CRM & Sales: HubSpot, Salesforce. Pull pipeline data, deal stages, conversion rates, or customer segments straight into a chart. No more waiting for your RevOps team to build a report.

• Marketing Automation & Analytics: Google Analytics, Mailchimp, ActiveCampaign, Klaviyo. Campaign performance, email metrics, audience data, all chartable without leaving Claude.

• Advertising & Social: Meta Ads Manager, Google Ads, LinkedIn Campaign Manager. Spend vs. return, audience breakdowns, campaign comparisons, visualized in seconds.

• Project Management: Asana, Monday.com, Jira. Track campaign timelines, team workloads, or sprint progress as visual dashboards.

• Communication & Collaboration: Slack, Gmail. Summarize and visualize patterns from conversations, feedback threads, or customer inquiries.

The workflow looks like this: Connect the tool. Ask Claude to pull the relevant data. Then ask for the chart. Three steps. One conversation. Real data.

👉 This means every chart in the list below can be built with your actual numbers, not hypothetical ones. That is the difference between a demo and a decision tool.

Subscribe now

Three Rules for Better Marketing Charts

Before you dive into the use cases below, three principles that make the difference between a mediocre chart and one that changes a meeting:

1. Be specific with your data. Vague prompts produce vague charts. The more concrete your numbers, labels, and dimensions, the better the output. Do not say “show me a funnel.” Say “show me a funnel with these five stages, these conversion rates, and these drop-off percentages.”

2. Always add context. Tell Claude who the chart is for and what decision it should support. A chart for your CEO looks different from a chart for your content team. Context shapes design.

3. Iterate in the same conversation. Your first chart will be 80% right. Stay in the session. Say “make the colors more distinct” or “add a trend line” or “change the X-axis to quarters.” Claude refines. You converge. That is how you work with an AI co-worker (and if you haven’t read my guide on that yet, start here).

👉 Premium subscribers unlock: How to share interactive charts with clients, stakeholders, and your team (public links, HTML exports, and presentation-ready formats). Plus all 10 ready-to-use marketing and sales use cases with copy-paste prompts: from performance dashboards and budget visualizations to funnel analysis, customer journey maps, competitive positioning, and marketing maturity assessments. Each with the exact prompt, a pro tip, and a live interactive example.

Read more

Everyone is talking about AI replacing jobs.

The panic version: millions of marketing roles will disappear overnight. The denial version: AI is just a tool, humans will always be in control.

Both are wrong.

And the reason both are wrong is the same: neither is based on data.

Last week, Anthropic published something genuinely rare in this debate. Not another thought piece. Not another prediction. A rigorous empirical study that measures what AI is actually doing in professional settings, right now, in real usage data (Massenkoff & McCrory, 2026).

The study covers the entire labor market. All occupations. From software developers to cashiers, from lawyers to agricultural workers.

I read it twice. And then I did what this newsletter exists to do: I took a broad, economy-wide study and ran it through a single lens.

👉 What does this mean specifically for marketing and sales?

Because buried inside a general labor market paper are findings that every CMO should be aware of. The strategic implications are more urgent than most organizations currently realize.

Why This Study Is Different

Most research on AI and labor markets measures what AI could theoretically do.

The standard approach, introduced by Eloundou et al. (2023), assigns each professional task a score based on whether an LLM could theoretically speed it up by at least 50%. This theoretical capability measure captures potential. But potential and reality are very different things.

Massenkoff & McCrory (2026) introduce something new: observed exposure.

The standard approach measures potential. Massenkoff & McCrory (2026) measure reality. They combine O*NET occupational data with real Claude usage data from the Anthropic Economic Index, weighting automated deployments more heavily than augmentative use, and work-related contexts more heavily than personal use.

The result is a measure that tracks the gap between capability and deployment.

This distinction matters enormously. Because right now, AI is far from reaching its theoretical ceiling. Actual coverage remains a fraction of what is feasible. The gap between blue and red in the Anthropic data is wide.

But that gap is not permanent. It is a trajectory.

Three Marketing & Sales Roles Under Pressure

This is where the data gets concrete for our profession.

Massenkoff & McCrory (2026) identify the ten most exposed occupations based on observed exposure. Three of them sit directly in marketing and sales.

Let me take each of these seriously.

1. Customer Service Representatives: 70.1% observed exposure

The leading automated task: Confer with customers to provide info, take orders, handle complaints.

This is not a marginal task. This is the core function of the role. And Massenkoff & McCrory (2026) note that Customer Service Representatives rank second among all occupations precisely because their primary tasks are increasingly visible in first-party API traffic. Organizations are already deploying AI here at scale.

For CMOs, the relevant question is no longer whether this is happening. It already is.

What matters is whether customer service automation gets treated as an HR and operations decision, or as a brand decision. Every automated customer interaction is a brand interaction. The design of those systems, the handoff logic, the moments where humans take over, these are brand architecture decisions that fall squarely in the CMO’s domain.

2. Market Research Analysts and Marketing Specialists: 64.8% observed exposure

The leading automated task: Prepare reports of findings, illustrating data graphically and translating complex findings into written text.

Read that sentence carefully.

The most exposed task for this role is the translation of findings into written reports and visualizations. Not data collection, not statistical modeling. The output layer. The thing most marketing analysts spend the largest share of their time actually producing.

When that layer migrates toward automation, the value of the role shifts toward hypothesis generation, strategic framing, and judgment calls that require context no report can summarize cleanly.

Organizations that understand this will redesign these roles proactively. Organizations that do not will find themselves with roles that feel redundant before anyone formally decides to eliminate them.

👉 I wrote about which marketing roles are fading and which are emerging in an earlier essay: The Future of Marketing Roles: Emerging and Fading Jobs.

Subscribe now

3. Sales Representatives: 62.8% observed exposure

The leading automated task: Contact customers to demonstrate products and solicit orders.

Outbound contact and product demonstration are high on the automation list. This does not mean that sales as a function disappears. It means that the entry-level and process-heavy components of the sales function are already being absorbed by AI systems in organizations that have chosen to deploy them.

The human value proposition in sales is shifting toward complexity resolution, relationship depth, and strategic account development. The transactional layer is leaving.

CMOs who own the marketing-sales interface need to factor this into how they think about pipeline architecture, lead handoff logic, and what kind of human touchpoints are worth investing in.

The Paradox: No Unemployment Spike, But a Hiring Freeze

Here is where the data gets genuinely surprising, and where intellectual honesty requires slowing down.

Massenkoff & McCrory (2026) find no statistically significant increase in unemployment among workers in the most exposed occupations since the release of ChatGPT in late 2022. None. The unemployment trends for exposed and unexposed workers have moved largely in parallel.

This is important. Anyone using this study to argue that marketing jobs are collapsing right now is misreading it.

But here is what the data also shows, and this is the part that matters most for CMOs thinking about talent architecture.

Among workers aged 22 to 25, there is a different story.

Massenkoff & McCrory (2026) find a roughly 14% decline in the job-finding rate for young workers entering highly exposed occupations, compared to the pre-ChatGPT baseline. The finding is barely statistically significant, but directionally consistent with Brynjolfsson et al. (2025), who report a 6 to 16% fall in employment in exposed occupations for the same age group.

The pipeline is quietly narrowing. Not through layoffs. Through a slowdown in the entry points. Young talent is being hired less frequently into the exposed roles. Whether because organizations need fewer junior positions, because the onboarding value of those roles has declined, or because the expected career trajectory has changed, the data does not say. But the direction is clear.

For CMOs, this is a talent pipeline signal, not a workforce crisis. The alarm is not ringing yet. But the pressure is building at the entry level.

And there is a second-order problem worth naming. If organizations stop hiring juniors into exposed roles, they also stop developing the next generation of senior marketers. The people who will eventually lead campaigns, run analytics teams, and own customer strategy are the juniors of today. Cut the entry point, and you cut the pipeline for the decade ahead. A profession that automates its own apprenticeship is making a bet it may regret.

Subscribe now

The Gap That Defines the Next Two Years

Perhaps the most important visual in the entire study is Figure 2.

Source: Massenkoff & McCrory (2026)

Massenkoff & McCrory (2026) state clearly: as capabilities advance, adoption spreads, and deployment deepens, the red area will grow to cover the blue.

This is the most important sentence in the study for marketing and sales leaders.

The current calm in the unemployment data does not reflect the long-run trajectory. It reflects the early state of deployment. The technology is ahead of the organizational adoption curve. That gap closes over time. And the organizations that treat that gap as a permanent cushion will find themselves unprepared when it narrows.

There is a parallel here to the history of digital transformation. In 2005, the companies that understood what the internet would do to retail had a significant planning advantage over those that waited for the disruption to become undeniable. The disruption was already visible in the data. It just had not fully materialized yet.

We are in a similar moment now, at the task level, inside professional roles.

Three Strategic Lenses for CMOs

What should marketing leaders actually do with this data?

Not panic. Not ignore it. Think structurally.

Lens 1: Role Design

Which tasks inside your exposed roles are being automated in practice, not just in theory? The Anthropic data specifies the leading automated task for each occupation. That is the place to start, and the answer should drive role redesign before it drives headcount decisions. What remains when the report-writing and outbound-contact layers are automated? What grows in value? That is where you invest in capability development.

One useful framework for thinking through this comes from earlier research on agentic marketing organizations. As execution migrates toward AI systems, three distinct human positions emerge:

Source: Own visualization based on Sukharevsky et al. (2025).

The pattern across all three positions is the same: AI handles the execution layer, while humans own the higher cognitive and socioemotional layers that machines cannot replicate. For CMOs redesigning exposed roles, this is the directional logic. A Market Research Analyst who no longer produces reports still has a role — but it looks more like the T-Shaped Specialist than the execution-heavy position it was. A customer service function that automates routine interactions still needs AI-Empowered Frontline Workers for the cases that require human judgment, empathy, and relationship depth.

👉 If you want to think through how to redesign marketing work structurally, two earlier essays go deeper: Task-Driven, Agentic Marketing Organizations and Designing Human Work in Agentic Marketing Organizations.

Lens 2: Talent Pipeline Architecture

If junior hiring into exposed roles is already slowing across the market, your organization is making an implicit choice whether you recognize it or not. Better to make it deliberately. What does your entry-level marketing and sales talent pipeline look like in three years if outbound contact, report generation, and data-translation tasks continue migrating toward automation? What capabilities need to enter at a different point in the career arc?

Lens 3: Customer Experience Ownership

Customer Service Representatives are the second most exposed occupation in the entire economy. They are also the primary touchpoints of your brand experience for most customers. If those roles are being automated in your organization or in your competitors’ organizations, the experience design questions that used to live in operations now belong to marketing leadership. Who decides where the AI stops and the human begins? Who owns the brand implications of that boundary? That is a CMO question.

👉 For a practical look at how marketing teams are already working with AI tools in their daily workflows, see: Claude Cowork for Marketing.

What the Data Actually Says

The Anthropic study does not tell us that marketing jobs will disappear.

It tells us something more specific and more useful: the tasks inside those jobs are already changing in the organizations that are deploying AI at scale. The unemployment data has not moved yet. But the hiring data for young workers has. And the theoretical capability that has not yet been deployed is still sitting in the gap between blue and red on that radar chart, waiting for the next capability jump, the next deployment cycle, the next organization that decides to move.

The companies that are designing for that future now, at the role level, at the talent pipeline level, at the brand experience level, will be in a fundamentally different position when the gap narrows.

The companies that wait for the unemployment data to tell them something has changed will be two years behind.

That is not a prediction. That is what the data already shows.

Yours,

Prof. Dr. Andreas Fuchs 🦊🎓

References

Brynjolfsson, E., Chandar, B., & Chen, R. (2025). Canaries in the coal mine? Six facts about the recent employment effects of artificial intelligence. Digital Economy.

Eloundou, T., Manning, S., Mishkin, P., & Rock, D. (2023). GPTs are GPTs: An early look at the labor market impact potential of large language models. arXiv. https://arxiv.org/abs/2303.10130

Massenkoff, M., & McCrory, P. (2026, March 5). Labor market impacts of AI: A new measure and early evidence. Anthropic. https://www.anthropic.com/research/labor-market-impacts

Part 1 (this guide): 20 Ways to Use AI as Your Marketing Co-Worker: How to set up Claude Cowork, how to work with it, and 20 use cases to start today.

Part 2: From Data to Decision in Minutes: How Claude Charts turns marketing data into interactive visualizations and dashboards.

Part 3: Skills are the New Skills in Marketing: How to create custom Skills that make Claude permanently better at your specific marketing work … with 10 ready-to-use Skill templates.

Be honest with yourself.

When was the last time you left the office thinking: “Today I did real marketing work”?

Not reformatting slides. Not rebuilding reports. Not rewriting the same briefing for the fourth time.

Real work. A decision that mattered. A strategy that moved something. Content you were proud of.

If you have to think about it, that is the problem.

Most marketers spend their weeks on repetitive tasks. Every document starts from scratch. Every report gets stitched together by hand. The same patterns, every week. All manual.

You know these tasks should be automated. There are tools for it. n8n. Make. Zapier. HubSpot workflows.

But nobody says this out loud: you were never trained for this. You studied marketing. Not workflow engineering. You know exactly which tasks eat your time. You just cannot automate them yourself.

And the AI tools you do use? You type a prompt. Get text back. Copy it. Paste it. Reformat it. Realize it does not match your voice, your brand, your context. Fix it by hand. That is not AI helping you. That is another task on the pile.

Two problems. No solution. Until now.

When Anthropic launched Cowork in January 2026, it wiped $285 billion off global software stocks in a single day. Not because of a new chatbot. Because Cowork lets marketers automate their tasks without building anything. No drag-and-drop. No API connectors. No technical setup. You describe the task. Cowork does it. Reads your files. Follows your brand voice. Produces real deliverables.

For the first time, the marketer who knows what needs to happen can make it happen. Without a developer. Without an automation specialist. Without learning tools that were never built for marketers.

This is an amplifier. It takes what you already know about marketing and multiplies your impact. Same thinking. Same standards. Same voice. But with a force behind it that one person could never have alone.

Every marketer needs to understand what this tool can do. This is the guide I wish someone had written for me when I started.

👉 How to set it up. How to use it right. And 20 use cases to start with today (read until end of article to see the full use cases incl. prompts).

What Is Claude Cowork? (And Why Should Marketers Care?)

Claude Cowork is not a chatbot. It is not another AI writing assistant. It is a work environment built into the Claude desktop app that changes how you interact with AI.

One thing it does not do: images. Cowork is built for text-based deliverables. Documents, spreadsheets, briefs, strategies, scripts, and even code. If you need visuals or graphics, use a dedicated image tool. I use Gemini for that. Know the boundary, use the right tool for the job.

Three things make it fundamentally different from what most marketers are used to.

It creates real files. Not just text.

You ask for a content calendar. You get an Excel file. Not text formatted to look like a table. An actual .xlsx file you can open, edit, and share.

You ask for a workshop handout. You get a Word document. Formatted. Structured. Ready to print.

You ask for a research brief. You get a markdown file. Organized by theme. With sources.

This sounds like a small thing. It is not. Because the biggest time killer in AI-assisted marketing work is not the generation. It is the reformatting. The copy-pasting. The manual transfer from “AI output” to “usable deliverable.”

Cowork eliminates that step entirely.

Subscribe now

It works with your context. Not generic templates.

This is where it gets interesting for marketers.

You can upload your brand guidelines, your writing style, your strategy documents, your past campaigns, your tone of voice rules. Cowork reads all of it. And it works within that context.

The output is not “AI-generated content.” It is content that sounds like you, follows your brand rules, and builds on your existing strategy.

Most AI tools start from zero every time. Cowork starts from where you are.

It works in sessions. Not single prompts.

Real marketing work is not a single question. It is a process.

You start with a brief. You get a first draft. You refine the angle. You adjust the tone. You add a section. You remove another. You pivot. You iterate.

With a typical chatbot, you lose the thread after 15 minutes. The AI starts repeating itself. Or contradicting what it said earlier. You open a new chat and start over.

Cowork handles multi-hour sessions. You can iterate, revise, and redirect without losing context. The AI remembers what you decided 60 minutes ago and builds on it.

For complex marketing tasks, this is the difference between “helpful” and “actually usable.”

It learns your workflows. Permanently.

This is where most people stop exploring. And where the real leverage begins.

Claude has a feature called Skills. A Skill is a set of instructions you write once that Claude follows from that point forward. Unlike a prompt that disappears when the conversation ends, a Skill persists. It lives in your account. It activates whenever the task matches.

Think about training a new team member. Day one, you explain how to write your reports. Day two, you explain it again. Day three, same thing. By week two, you are exhausted.

A Skill is writing it down once. Claude reads it, follows it, delivers the work exactly how you want. Every time.

The context files I describe in the setup section below (voice profile, brand context, working preferences) are the first step. They give Claude your context for a single session. Skills go further. They make your workflows permanent, reusable, and stackable.

For example: you can create a Skill for your monthly reporting format. Another Skill for your content approval process. Another for how you brief agencies. Each one fires automatically when the task matches. No re-explaining. No re-prompting.

And here is the part that matters for teams: Skills are shareable. You build one, your entire marketing team uses it. Same quality. Same format. Same voice. Every time.

You do not need to build Skills from scratch. Claude has a built-in skill-creator. Just type: “I want to create a Skill for [your use case].” Claude asks about your process, you answer naturally, and it generates the Skill file automatically.

Try this: “I want to create a Skill for my monthly reporting. Ask me about my process, then generate the Skill file.”

If you already have a conversation where Claude nailed the output, type “Create a skill to remember this.” One sentence turns a good conversation into a permanent workflow. You can also browse and install ready-made Skills from Anthropic’s official Skills library. Find one that matches your workflow, install it, done.

👉 Start with context files. Graduate to Skills once you see what Cowork can do.

How to Get Started. Step by Step.

Setting up Cowork takes about 10 minutes. But setting it up right takes a bit more thought. Here is the process I recommend.

Step 1: Install Claude and Open Cowork

Download the Claude desktop app from claude.ai/download (it does not work in your browser). You need a Pro account ($20/month). Once installed, open Claude and switch to the “Cowork” tab at the top.

That is it. You are in.

Step 2: Configure Three Settings

Before you do anything else, turn on three settings that make everything work better. Most people skip this. Do not.

1. Memory (Settings > Personalisation > Enable Memory). Not on by default. Once enabled, Claude remembers your preferences across all conversations. It learns how you work over time. Without this, every new chat starts from zero.

   

2. Extended Thinking (Enable in settings or per message). Claude reasons step-by-step before answering. You see its thinking process in grey text. It uses more tokens but produces significantly better output for complex tasks like strategy documents, research briefs, and multi-step workflows.

   

3. Projects (Available in the sidebar). This is where it gets powerful for marketers. Create a Project for each client, campaign, or content series. Upload files as persistent context. Add project-specific instructions. Every conversation inside the project shares the same context. Your brand guidelines, your strategy documents, your voice profile. All of it is always available. No re-uploading. No re-explaining.

👉 Memory works across Projects. Your preferences follow you everywhere. Projects keep your marketing work organized and your context persistent.

Step 3: Build Your Marketing Context Files

This is the most important step. And the one most people skip.

Before you start your first real task, create a set of context files. Think of them as your onboarding package for a new team member. The better the briefing, the better the output.

Here is what I recommend:

File 1: Voice Profile (voice-profile.md)

This file describes how you write. Not what you write about. How.

Include things like: your typical sentence length. Whether you use short punchy openers or longer introductions. Words and phrases you always use. Words and phrases you never use. Whether you use emojis and which ones. How you structure arguments. How you close articles or posts.

The more specific, the better. “Professional but approachable” is useless. “Short sentences for key points. Longer sentences for explanations. Never use em dashes. Always end with 👉 for key takeaways.” That is useful.

👉 Pro tip: take 3 of your best pieces of content and reverse-engineer the patterns. What makes them sound like you?

👉 Not sure how to create a markdown file? You don’t have to. Ask Cowork: “Read these 3 articles and create a voice-profile.md file based on them.” It builds the file for you. That is the whole point. And if you have been using ChatGPT or another AI tool before, you can ask it to create the markdown file for you there and then upload it to Cowork.

Download Voice-Profile.md (Premium Only)

File 2: Brand Context (brand-context.md)

This file describes your brand. Your company’s positioning. Your target audience. Your value proposition. Your key messages. Your competitive landscape.

If you have an existing brand book or positioning document, you can upload that directly. If you don’t, create a simple markdown file with the essentials.

Download Brand-Context-md (Premium Only)

File 3: Working Preferences (preferences.md)

This file tells Cowork how you like to work. Do you want it to ask clarifying questions before starting? Or just deliver a first draft? Do you prefer bullet points or prose? What level of detail do you expect? What file formats do you want?

This might feel unnecessary. It is not. It saves you 5-10 rounds of back-and-forth corrections in every single session.

Download Preferences.md (Premium Only)

Step 4: Start With a Real Task

Do not test Cowork with a toy problem. Do not ask it “What can you do?” Start with something you actually need done this week.

Open Cowork. Upload your context files. Give it a clear brief. See what happens.

Your first prompt: “Here are my context files. Read them carefully. Then write a first draft for a LinkedIn post about [your topic]. Match my voice profile. Save it as a markdown file.”

The best way to learn how Cowork works is to use it for real work. Pick one task from the list at the end of this essay and run it.

Three Principles for Working With Cowork

Setting up the tool is the easy part. Working with it effectively requires a shift in how you think about AI collaboration.

Here are three principles that made the biggest difference for me.

Principle 1: Brief it like a smart junior marketer.

Most people prompt AI like a search engine. Short queries. Vague instructions. Then they complain the output is generic.

Cowork responds to the quality of your briefing. Treat it like you would treat a talented but new team member who needs clear direction.

Not: “Write me a blog post about data strategy.”

But: “Here is my latest article about data strategy [uploaded]. Here is my audience: marketing leaders in B2B companies. I want a LinkedIn post that opens with a counterintuitive statement, explains one key framework in 3 sentences, and ends with a question. 150 words max. Match the tone from my voice profile.”

The more specific, the better. Cowork does not mind long briefs. It actually performs better with them.

Principle 2: Iterate in the same session.

Do not start a new conversation for every revision. Stay in the session. Build on what you have.

Say things like: “The intro is too long. Cut it in half. Keep the closing.” Or: “Good structure, but the tone feels like a press release. Make it sound like a conversation.”

Cowork gets better within a session because it accumulates context from your feedback. Every correction teaches it what you want. Leaving the session and starting fresh means losing all of that context.

👉 Think of a Cowork session like a working meeting, not a series of separate requests.

Principle 3: Always ask for files. Not just text.

Whenever possible, ask for a deliverable. Not text in the chat window. A file.

“Create an Excel file with...” “Give me a Word document that...” “Produce a markdown brief with...”

Files are actionable. Chat text is not. Files can be shared with your team, opened in other tools, and iterated on outside of Cowork. That is the whole point of a co-worker: they produce deliverables.

For Advanced Users: Four Power Features

Below for Premium subscribers: 20 ready-to-use prompts for content strategy, campaign planning, competitor analysis, reporting, and more. Each with the exact prompt, expected output, and tips.

Read more

This is a practical deep dive from Future of Marketing. It is based on my own experience growing this Substack to over 1,000 subscribers in a few months, using LinkedIn as the primary distribution channel.

Most Substack writers face the same quiet frustration. They write thoughtful pieces. They publish consistently. They invest hours into research, structure, and clarity. And then … almost nothing happens.

The reason is not the quality of the writing. It is the absence of distribution.

Substack rewards depth, not reach. It builds relationships, not audiences. That is its strength — but also its limitation for anyone starting out. The question is not how do I write better? It is: how do I get the right people to discover that I write at all?

For me, the answer has been LinkedIn. Not as a promotional channel. Not as a place to drop links. But as a reach lever — a system that creates visibility, builds trust, and converts attention into subscribers. With nearly 6,000 followers on LinkedIn, the platform has become my most important distribution channel — not because of the follower count itself, but because of the system behind it.

In this playbook, I share what I have learned. Step by step. With real examples, real numbers, and real principles.

👉 Read this article to the end.

Why LinkedIn? The Logic Behind the Lever

LinkedIn is not social media in the traditional sense. It is a professional attention market. People come with an intent to learn, to stay current, to signal competence. That creates a natural alignment with what Substack offers: long-form, expert-driven content for professionals who want to go deeper.

LinkedIn’s algorithm amplifies content that generates engagement — comments, dwell time, conversation. But 1,300 characters in a feed post is not enough to build a real argument. That is where Substack takes over. Your LinkedIn profile becomes the bridge. Your posts become the invitations. Your Substack becomes the destination.

👉 Think of LinkedIn as the storefront and Substack as the workshop. One catches attention. The other earns trust.

Step 1: Optimize Your LinkedIn Profile as a Conversion Path

Your LinkedIn profile is not a resume. It is a landing page.

Every element should answer one question: Why should someone subscribe to your Substack?

Most profiles fail here because they try to communicate everything — job history, skills, endorsements, publications. But for Substack growth, your profile needs to do one thing well: make the path from curiosity to subscription as short as possible.

Here is a mini checklist you can apply immediately:

• Banner: Use one clear message and one clear CTA. Your Substack link should be visible at first glance. Do not clutter the banner with logos, titles, and decorative elements. One message. One action.

• Headline: State who it is for and what transformation you offer. Not your title. Your promise. My headline, for example, does not just say “Professor of Marketing.” It says what I do and why it matters.

• About / Info section: This is where proof, offer, and link come together. Lead with credibility, explain what your Substack delivers, and make the link easy to find. Do not bury it at the bottom.

• Featured section: Pin your best-performing post and your Substack landing page. This is prime real estate — use it.

• Premium / Newsletter button: LinkedIn Premium allows you to add a custom button that links directly to your Substack. This is your primary conversion path. Make it point to your Substack, not your website.

• Consistency check: The same promise should appear across banner, headline, and About section. No mixed messages. If your banner says “Future of Marketing” but your headline talks about consulting, you create friction.

👉 Your profile is not about you. It is about the person visiting it and the question they carry: Is this worth my attention?

Check Out My LinkedIn For Best Practice

Step 2: Create Content That Earns Reach

LinkedIn’s algorithm is not a mystery. It follows a set of principles that, once understood, can be leveraged deliberately. Not to game the system — but to ensure that good content actually reaches the people it was written for.

Here are the five algorithm principles that matter most:

1. Early engagement matters. Strong activity in the first one to two hours after posting increases distribution significantly. The algorithm tests your content with a small initial audience and then decides whether to expand its reach. What to do: reply fast to comments, ask follow-up questions, and keep conversation threads alive.

2. Dwell time and depth. Posts that people actually read — not just scroll past — tend to be shown to more people. What to do: write for clarity, use structure, and make the first lines irresistible. The hook is not a trick. It is a promise.

3. Comments carry more weight than likes. A like is a nod. A comment is a conversation. LinkedIn treats comments as a deeper engagement signal. What to do: invite opinions, trade-offs, and real experiences. Ask questions that are worth answering.

4. Network relevance. LinkedIn starts by testing your content within your own network before expanding beyond it. What to do: speak to a clearly defined audience and publish on consistent topics. If your content is about marketing one day and cryptocurrency the next, the algorithm cannot find your audience.

5. Conversation loops. Every reply to a comment keeps the post “alive” in the feed. What to do: respond with genuine follow-up questions, tag people when relevant, and treat the comment section as an extension of the post itself.

The Do / Don’t principle is simple:

• Do: reply within the first hour, turn comments into discussions, and be specific.

• Don’t: drop links too early, ignore comments, or write generic takes that could apply to any industry.

👉 The algorithm does not reward cleverness. It rewards engagement. And engagement starts with relevance.

Premium subscribers unlock: The exact lead magnet strategy that generated 190,000 impressions and 2,100 comments from a single post. The launch campaign sequence I used (with real screenshots). My direct outreach system for targeted subscriber conversion. And the full metrics breakdown showing which tactics drove the most subscribers per hour invested.

Read more

Since February 9, 2026, ChatGPT shows ads.

For now, it is a test. Limited to the United States. Visible only to users on the Free and Go tiers. Ads appear at the bottom of responses, visually separated, clearly labeled as sponsored. Those who pay for Plus, Pro, or Enterprise see nothing.

On the surface, this looks harmless. A familiar pattern: free product, rising costs, advertising as the financing model. Half the internet runs on this logic.

But if you look more carefully, something else is happening.

The Intimacy Problem of AI Chatbots

ChatGPT is not a search result. Not a content channel. Not a social media feed. It is a conversational partner.

People ask it questions they would otherwise ask a consultant. Or a doctor. Or no one at all. The psychological proximity between user and interface is fundamentally different from what we know from Google, Instagram, or YouTube.

Scott Galloway, marketing professor at NYU Stern, made this point with striking clarity after Anthropic’s Super Bowl campaign aired. What makes advertising in AI so different, he argued, is that the dominant use case for ChatGPT is not productivity. It is something closer to therapy. Users confide in it. They share vulnerabilities. They ask for help with relationships, mental health, life decisions.

Now imagine that moment being interrupted by a sponsored message.

OpenAI insists this will not happen. Ads will not appear near sensitive topics, they say. Conversations remain private. Advertisers receive only aggregate data. The model itself does not know when an ad is present.

These are reasonable guardrails. But they are also exactly the kinds of promises every platform has made at the beginning … and abandoned later.

The Pattern We Have Seen Before

Sam Altman himself, speaking at Harvard University in May 2024, called the combination of ads and AI “uniquely unsettling” and described advertising as a “last resort” for OpenAI’s business model. By July 2025, his tone had softened. By January 2026, ads were announced. By February 2026, they were live.

This is not evolution. This is capitulation before the business model that already reshaped the internet once.

Cory Doctorow, author, activist, and one of the sharpest critics of digital platform economics, coined a term for this trajectory: Enshittification. The pattern goes like this: attract users with quality. Monetize their attention. Then gradually optimize for extraction — at the expense of the experience that attracted them in the first place.

Google took this path. Facebook did too. The promises at the beginning sounded remarkably similar to what OpenAI is saying today.

👉 The question is not whether OpenAI means well. The question is whether the incentive structure that advertising creates will gradually shift what “helpful” means inside the system.

Because once a platform earns revenue from showing ads, it faces a structural tension: serve the user or serve the advertiser? Every platform in history has resolved this tension in the same direction. Not because the people running them were dishonest. But because the economics made it inevitable.

The New Fault Line: Ad Platform or Trusted Advisor?

What makes this moment particularly interesting from a marketing perspective is that the competitive landscape is now dividing along the advertising question.

Anthropic built an entire Super Bowl campaign around the promise of keeping Claude ad-free. Four spots. Millions of viewers. The tagline: “Ads are coming to AI. But not to Claude.” The ads were satirical, sharp, and effective. They showed fictional chatbots interrupting intimate conversations with absurd sponsored messages — a mature dating site during a therapy session, height-boosting insoles during a fitness consultation.

Mark Ritson, one of the most respected voices in brand strategy, called it the first piece of effective brand strategy the AI category has produced.

Sam Altman responded publicly and called the campaign “dishonest” and Anthropic “authoritarian.” The intensity of his reaction tells you something: Anthropic hit a nerve.

Google has signaled that Gemini could follow a similar path toward ad integration. Perplexity has already experimented with sponsored placements in its AI search results.

A new fault line is emerging in the AI industry: Who becomes an advertising platform? And who deliberately does not?

This is not a technical distinction. It is a strategic one. And it will shape which AI systems marketers trust, which ones consumers trust, and which ones ultimately mediate the relationship between brands and buyers.

Subscribe now

What This Means for Marketers

For marketing professionals, the immediate implication is clear: a new advertising channel is opening. ChatGPT has over 800 million weekly users with high intent. OpenAI has already onboarded major advertisers through its Ad Pilot Program. Omnicom Media has secured placements for more than 30 clients across verticals from automotive to retail. Adobe is running pilot ads for Acrobat Studio and Adobe Firefly. The minimum commitment is reportedly $200,000.

This is relevant. And it would be naive to ignore it.

But the deeper question is strategic, not tactical.

If AI interfaces increasingly mediate product discovery, comparison, and purchase decisions, then the boundary between “helpful answer” and “advertisement” becomes the most consequential line in digital marketing. And the entity drawing that line is not a regulator or an industry body. It is the AI company itself.

👉 Today, OpenAI says ads do not influence ChatGPT’s answers. But the question is not about today. It is about what happens when advertising revenue becomes 10%, 20%, 30% of OpenAI’s income. When optimizing ad relevance means optimizing conversation data. When “helpful” and “commercially valuable” gradually converge until no one can tell the difference anymore.

We have seen this movie before. We know how it ends.

When Monetization Enters Cognitive Space

The issue is not whether ChatGPT displays a banner ad at the bottom of a response.

The issue is what happens when the most trusted interface of our time adopts the structural logic of the advertising market.

Traditional advertising operates in contexts where people expect to be sold to: a television commercial break, a billboard, a sponsored search result. The social contract is transparent. You know you are in a commercial environment. Your cognitive defenses are active.

AI conversations are different. They operate in a space of cognitive intimacy. Users lower their defenses. They share context they would not share with a search engine. They accept recommendations with a trust level that resembles personal advice, not commercial suggestion.

When monetization enters this cognitive space, neutrality stops being a feature. It becomes a governance problem.

Who audits whether an AI’s recommendations are genuinely optimal for the user or subtly shaped by commercial incentives? Who defines the threshold between “relevant sponsored content” and “manipulative placement”? Who ensures that the wall between organic answers and advertising holds — not just at launch, but five years from now, when revenue pressure intensifies?

These are not hypothetical questions. They are the defining regulatory and strategic questions of the next decade in marketing.

Where This Leaves Us

OpenAI’s move to introduce advertising in ChatGPT is not surprising. It follows a financial logic that is difficult to argue with. Running large language models at scale is extraordinarily expensive. Subscription revenue alone may not cover the costs. Advertising offers a path to sustainability.

But sustainability and integrity are not the same thing.

The platforms that shaped the last two decades of digital marketing … Google, Facebook, Amazon … all started with promises about user-first design. They all built advertising models that gradually shifted incentives. They all ended up in a place where user trust became a cost of doing business rather than a foundation for it.

The AI industry now faces the same fork in the road. And the decisions being made right now — by OpenAI, by Anthropic, by Google, and by the marketers who allocate budgets to these channels — will determine whether AI interfaces become the next generation of trusted advisors or the next generation of ad platforms.

We should watch very closely.

Yours,

Prof. Dr. Andreas Fuchs 🦊🎓

In my last post, I outlined the CMO Agenda 2026 that will shape this Substack over the coming weeks and months.

Based on the State of Marketing Europe 2026 report, we identified the ten themes (the CMO Agenda 2026) that currently matter most to CMOs across Europe (see illustration below).

Today, we begin with an unusual starting point.

Not with the topic ranked #1. But with the one ranked last.

#10: Data-Driven Marketing.

And I want to argue that this ranking tells us something far more important than it seems.

Why Last Place Is Not What You Think

At first glance, you might read “#10” as a demotion. A signal that data has lost its relevance. That CMOs have moved on.

That would be the wrong conclusion.

When 500 senior marketing leaders across Europe rank data-driven marketing at the bottom of the top ten, they are not saying it does not matter. They are saying something more nuanced:

Data has become assumed infrastructure.

It is no longer a strategic initiative. It is a precondition.

Think about it this way: Nobody puts “electricity” on a list of strategic priorities either. Not because it is irrelevant — but because without it, nothing else on the list works.

Branding (#1)? Requires data to measure distinctiveness, track perception, and allocate budgets. Marketing ROI (#6)? Impossible without clean, shared data. Connecting marketing and sales (#7)? A coordination problem that collapses without a common data foundation. Personalization (#12)? Entirely data-dependent.

Data does not compete with these themes. It runs beneath all of them.

And that is exactly why its position at #10 is not a weakness — it is a sign of maturation.

The Paradox: DDM Is Everywhere Present, But Nowhere Mastered

But here is the tension.

If data-driven marketing has become foundational, you might expect organisations to have figured it out by now. They have not.

McKinsey’s report classifies data-driven marketing as a medium need-for-action topic — not the highest urgency, but far from resolved. The gap between perceived importance and actual maturity remains real.

And in every organisation I work with — whether mid-market or enterprise, B2B or B2C — I see the same pattern:

More data than ever. Less shared understanding than ever.

Dashboards multiply. Reports circulate. Analytics teams grow. And yet, in leadership meetings, people still argue about which numbers to trust.

This is the paradox at the centre of modern marketing:

We have solved the access problem. We have not solved the alignment problem.

Most organisations today do not struggle to collect data. They struggle to agree on what the data means — and what to do with it.

Subscribe now

From Reporting Layer to Coordination Mechanism

This is where the conversation needs to shift.

For the past decade, data-driven marketing has been framed as a reporting function. Build the dashboard. Track the KPIs. Report to leadership. The implicit assumption: once people see the numbers, they will make better decisions.

That assumption is wrong.

Data does not create alignment by being visible. It creates alignment by being trusted, shared, and decision-relevant.

The real question is not: “Do we have the data?”

The real question is: “Do we agree on what the data tells us — and does it change what we do?”

This reframing matters enormously. Because it shifts data-driven marketing from a technical capability to an organisational coordination mechanism.

And if you have been reading this Substack, you know that this is not a new thought. It connects directly to the Marketing Operating System (MOS) I introduced earlier. The MOS is built on three layers — Strategy, Process, and Insights — and runs through four dynamic phases: Know, Analyze, Orchestrate, and Engage.

Data-driven marketing is not one of those phases. It is the thread that connects all of them.

Source: Fuchs (2025).

Without a functioning data foundation, the KNOW phase has no memory. The ANALYZE phase has no inputs. The ORCHESTRATE phase has no signals. And the ENGAGE phase has no feedback loop.

Data is the nervous system of the entire Marketing OS. When it works, the system learns. When it breaks, the system guesses.

Let me unpack what this means in practice.

A Playbook of Four Shifts That Defines the New Data-Driven Marketing

1. From Dashboards to Shared Reality

Read more

Marketing is moving faster than most organizations can adapt. My mission with Future•of•Marketing has always been to give you the clarity needed to stay ahead. To deliver the depth and rigor your marketing transformation requires. I want this newsletter to remain the best marketing investment of your the week, whether you are a marketing decider, a CMO or a marketing practitioner.

Today, with around 1,000 subscribers and almost the same number of followers reading along, I’d like to share why I’m taking the next step with Future of Marketing.

✨ Why Future of Marketing goes Premium

Over time, this Substack has evolved. The texts have become longer, more analytical, and more demanding. Both to write and to read. What began as public reflection has increasingly turned into a space for deeper thinking. Essays that take time. Arguments that resist simplification. Ideas that are not optimized for speed or reach.

At some point, the existing structure no longer fit that kind of work.

Moving to a Premium edition is not about restricting access or commercializing the publication. It is about creating a framework that protects focus and independence. Writing that remains fully free is inevitably shaped by immediacy and attention. A paid structure creates the distance needed for ideas to develop with care and depth.

This shift is also about depth over reach. I am not writing for likes or algorithms, but for people who want to build and shape the future of marketing. I would rather publish one essay that remains relevant for a year than several short takes that disappear within days. For readers, this means less noise and more signal. Fewer trends to chase, more frameworks that actually inform decisions.

Finally, it changes the relationship between writer and reader. When you invest in your knowledge, I invest more deeply in research, synthesis, and argumentation. You read with greater focus. I write with greater rigor. This shared commitment raises the quality of the work on both sides.

👉 The Premium edition will be available starting February 9.

Support FOM & Go Premium

What Premium will change for you?

I will continue to provide free weekly perspectives, but the Full Experience is moving to the Premium tier to support the deeper research and high-quality production this field demands.

While the free tier remains your window into these topics, the Premium tier is where the actual transformation happens. It is the exclusive home for the templates, guides, and playbooks you need to move from merely understanding a concept to successfully implementing a strategy.

Free Subscribers get:

• 💡 Weekly perspectives to keep you informed about the most relevant topics shaping the future of marketing in under 15 Minutes.

Premium Subscribers unlock:

• 📘 2–3 marketing deep dives per month:

  Comprehensive essays, guides, and workshops on the most critical topics.

• 📚 Full access to the premium content library:

  All essays, frameworks, workshops & long-form analyses

• 🧰 Curated premium resources:

  Selected case studies, playbooks, templates for practitioners

• 🧭 Orientation: Where marketing is heading and what this means for your strategy, organization, and daily decisions.

• 🎓 Highest premium quality

  Built on academic rigor and real-world practice. Research-backed, experience-tested, decision-ready. By Prof. Dr. Fuchs. Leading voice in modern marketing.

Support FOM & Go Premium

Quality is a choice. Thank you for choosing to go deep. See you next week.

Yours,

Prof. Dr. Andreas Fuchs 🦊🎓

I’ve been quiet for three weeks.

Not because the world slowed down.

But because mine did.

And nothing recalibrates your sense of relevance, urgency, and signal versus noise more effectively than that.

When you return after such a pause, one thing becomes painfully obvious:

Most debates in marketing are far too narrow for the moment we are living in.

The Illusion of AI as the Single Big Theme

If you’ve followed my recent writing, you might think everything currently revolves around agentic marketing: AI agents. Task-driven organizations. Machines mediating choice. All of that matters. Deeply.

But it is not the whole story.

Marketing leaders are not obsessed with a single technology shift. They are navigating a stack of simultaneous tensions:

• trust and efficiency

• brand and ROI

• data and privacy

• speed and meaning.

And this becomes visible only when you step back from individual essays and consider the broader pattern they form.

👉 Read this article to the end.

What CMOs Are Actually Thinking About

McKinsey’s State of Marketing Europe 2026 report offers a rare empirical snapshot of this reality. Based on interviews with 500 senior marketing leaders across Europe, it doesn’t ask what should matter, but what does.

What follows is not just an overview of topics, but a preview of what this Substack (Future of Marketing) will explore in depth over the coming weeks and months. Each of these themes will be unpacked in dedicated deep dives here on Substack. One topic at a time. One structural lens at a time.

Top 10: The Core Agenda of the next months

A preview of the deep dives to come:

1. Branding: Branding has returned to the top not despite technological acceleration, but because of it. As interfaces, platforms, and algorithms increasingly mediate choice, brand becomes the last stable orientation point for customers. In the coming deep dive, we’ll look at branding as long-term infrastructure: how emotional relevance, distinctiveness, and consistency function as economic assets in volatile markets, and why CMOs are doubling down on brand building precisely when pressure to cut grows strongest.

2. Budget Management: Budget management is no longer an operational concern; it is a governance question. CMOs are expected to allocate capital with CFO-level rigor while still defending long-term investments that don’t pay back within a quarter. This topic will explore how marketing leaders balance short-term efficiency with through-cycle investment logic, and why disciplined allocation is now inseparable from strategic credibility in the C-suite.

3. Data Privacy: Privacy has evolved from compliance to competitive signal. In Europe especially, how a company governs data has become inseparable from how much it is trusted. The deep dive will examine why privacy failures destroy brand equity faster than failed campaigns, how consent and governance shape customer relationships, and why trust is increasingly encoded in systems rather than statements.

4. Authenticity: Authenticity is no longer about tone of voice or storytelling techniques. In a world of synthetic content and AI-generated media, credibility emerges from coherence over time. This section will explore authenticity as a systemic property: how transparency, consistency, and proof replace performance, and why audiences now infer truth rather than accept claims.

5. Employer Branding: Talent markets have become belief markets. Employer branding now mirrors consumer branding, driven by purpose, culture, and credibility rather than compensation alone. What has changed fundamentally is ownership: employer branding is no longer an HR topic, but a marketing one. Candidates trust people more than promises, and employees have become the most credible media channel an organization has. As internal and external brand narratives collapse into one, corporate influencers play a growing role in shaping employer reputation through lived experience rather than managed messaging.

6. Marketing ROI: Marketing ROI sits at the intersection of legitimacy and survival. Boards demand proof, yet traditional attribution models increasingly fail in privacy-constrained, multi-touch environments. This deep dive will unpack why CMOs are shifting toward probabilistic measurement, incrementality, and holistic models, and how ROI is becoming a shared language between marketing, finance, and strategy.

7. Connecting Marketing and Sales: Growth does not happen in departments; it happens in systems. The persistent gap between marketing and sales is no longer an inefficiency but a structural risk. We’ll explore how leading organizations are dissolving functional boundaries, aligning incentives, and redesigning operating models so that demand generation and conversion operate as one continuous mechanism.

8. Integrating Marketing and Customer Experience: The brand promise is no longer communicated; it is executed. Customer experience has become the most powerful brand channel, whether organizations intend it or not. This topic will examine how CMOs increasingly take custody of the end-to-end journey, why experience consistency matters more than message consistency, and how systems deliver meaning repeatedly or not at all.

9. Agile Working: Agility has moved from methodology to survival trait. Annual plans fracture under volatility, and static hierarchies struggle with exponential change. We’ll look at how agile ways of working reshape decision rights, speed, and accountability, and why cultural change, not process theater, determines whether agility creates clarity or chaos.

10. Data-Driven Marketing: Data-driven marketing is no longer about dashboards. It is about shared reality. As organizations scale analytics, the challenge shifts from access to alignment: which numbers matter, who trusts them, and how decisions are made. This deep dive will explore data as a coordination mechanism across teams rather than a reporting layer.

    Subscribe now

Top 11–20: The Orbiting Forces

A preview of the extended list of deep dives to come in the next months:

11. Community Building: Community is re-emerging as a counterweight to fragmentation and algorithmic mediation. Beyond reach and engagement, communities create belonging, resilience, and advocacy. We’ll explore why CMOs increasingly invest in two-way relationships, how communities function as trust reservoirs, and why they resist short-term optimization logic.

12. Personalization: Personalization is moving from rules to systems. With GenAI, one-to-one experiences become technically feasible, but organizationally complex. This topic will examine where personalization truly creates value, how privacy and relevance interact, and why poorly governed personalization erodes trust faster than generic communication.

13. Creative Content: Creativity has quietly slipped down the ranking, yet its importance has not diminished. In oversaturated environments, creativity is the only sustainable differentiator. We’ll look at why creativity is being reframed as an efficiency lever, how GenAI raises the creative floor but not the ceiling, and why human judgment remains irreplaceable.

14. Sustainability-focused Marketing: Sustainability is shifting from positioning to proof. As greenwashing fatigue rises, only operational substance survives scrutiny. This deep dive will explore how sustainability claims intersect with trust, regulation, and long-term brand value, and why CMOs must increasingly align messaging with verifiable impact.

15. Channel Mix: Channel strategy is becoming fluid, not optimized. As discovery fragments across platforms, creators, and AI-driven interfaces, channel mix decisions turn into portfolio bets under uncertainty. We’ll explore how CMOs balance reach, attention quality, and control, and why channel strategy is increasingly revisited rather than fixed.

16. Macro Developments: Marketing does not operate outside history. Geopolitics, inflation, demographic shifts, and technological power concentrations shape consumer behavior long before campaigns do. This topic will connect macro uncertainty to brand trust, pricing power, and investment logic, highlighting why CMOs are becoming students of systems again.

17. GenAI and Agents: GenAI ranks low in priority but high in urgency. This contradiction defines the current moment. We’ll examine why many CMOs underestimate AI’s structural impact, how efficiency gains translate into strategic advantage, and why agentic systems represent a qualitative shift from automation to delegation.

18. MarTech and AdTech: The tech stack has become both enabler and bottleneck. Fragmentation, integration debt, and capability gaps limit value realization. This deep dive will focus on MarTech not as tooling, but as operating infrastructure, and why architecture decisions increasingly shape organizational speed and optionality.

19. Commerce Media: Commerce media blurs the line between marketing, sales, and platform economics. As retailers become media owners, attention turns into inventory. We’ll explore how this reshapes power dynamics, measurement logic, and the role of marketing within revenue engines.

20. Social Media and Influencers: Social platforms are no longer distribution channels; they are cultural arenas. Influencers function as trust proxies in fragmented attention economies. This topic will examine how credibility is borrowed, lost, or compounded through creators, and why social media increasingly reflects identity rather than promotion.

What Comes Next on This Substack

This essay is not a conclusion. It is a starting point.

The agenda outlined above is not about predicting the next tool or tactic. It is about understanding the pressures, trade-offs, and contradictions that define marketing leadership right now. Between growth and trust. Between efficiency and meaning. Between automation and responsibility.

Over the coming weeks and months, I will take each of these themes and explore them here on Substack. One at a time. In depth. Not as isolated trends, but as interconnected forces reshaping how marketing is practiced, governed, and judged.

This work is written for CMOs and senior marketing decision-makers who carry responsibility not just for performance, but for direction. For those who sit between the boardroom and the market, between strategy and execution, and who feel the weight of choices that compound over time.

Because marketing does not change when technology changes.

It changes when its underlying assumptions no longer fit reality.

This Substack is my attempt to examine that shift with distance rather than urgency, with structure rather than noise, and with a longer time horizon than the next quarter allows.

If marketing is becoming less about persuasion and more about coordination, then clarity, judgment, and restraint will matter more than speed.

That is the work ahead.

Yours,

Prof. Dr. Andreas Fuchs 🦊🎓

Marketing organizations are moving away from role-based hierarchies toward task-driven, agentic systems, where humans and AI agents collaborate at scale.

This shift changes how work is executed.

But it raises a more fundamental question:

This essay is not an attempt to name new roles for an AI age.

It is an attempt to rethink where human contribution belongs once execution itself becomes scalable.

Why Roles Lose Their Explanatory Power

Roles are an organizational invention optimized for scarcity.

• They bundle tasks because human time is limited.

• They stabilize accountability when coordination is costly.

• They make organizations manageable when execution depends primarily on people.

Agentic systems quietly dissolve these assumptions.

Tasks can now be decomposed, recombined, and executed by specialized agents at near-zero marginal cost. Execution no longer scales with headcount. Work becomes continuous, parallel, and fluid.

Under these conditions, roles do not suddenly disappear.

• They simply stop explaining how value is created.

• They turn into a legacy interface between humans and work.

The instinctive response is familiar. New titles appear. AI managers. Automation leads. Prompt engineers.

But changing labels preserves the logic that is already failing.

The real transition is not from old roles to new ones.

It is from roles to the deliberate design of human work.

From Managing People to Designing Work

Once execution is delegated to agents, leadership changes its nature.

The central question is no longer:

Who is responsible for what?

It becomes:

Where should humans be positioned relative to workflows, agents, and outcomes?

As machines take over execution, human value moves upward.

Humans stop being the primary executors of tasks. They become owners, designers, and stewards of systems. Human work can no longer be inherited from org charts designed for a different era. It has to be designed intentionally.

The Three Human Positions in Agentic Marketing

What emerges is not a new org chart, but a set of recurring human positions. Structural positions that agentic systems require in order to function without drifting into chaos.

Source: Own visualization based on Sukharevsky et al. (2025).

1. Orchestrators of End-to-End Work

(M-Shaped General Manager)

Orchestrators represent the evolution of management when managing people is no longer the bottleneck.

Their responsibility is not supervision. It is system design.

They define outcomes, decompose work into workflows, and decide where automation operates autonomously and where human involvement remains essential. They manage trade-offs between speed, quality, risk, and cost.

Crucially, they retain accountability.

Execution can be automated. Accountability cannot. It must remain human.

Orchestrators do not manage teams or channels.

They orchestrate interaction between capabilities.

Example: A Head of Growth who no longer runs channel teams, but owns end-to-end revenue workflows across marketing, sales, and AI agents and is accountable for outcomes rather than headcount.

Skill gap: Compared to today’s manager roles, this profile requires a novel mix of integrative thinking and the ability to manage AI-first workflows.

2. Stewards of Expertise and Exceptions

(T-Shaped Deep Specialist)

As automation scales, deep expertise becomes more valuable, not less.

Stewards of expertise provide human oversight in agentic systems. They fine-tune and teach agents, handle exception cases, and protect quality where automated execution alone is insufficient.

They combine deep domain knowledge with higher cognitive abilities such as pattern recognition, judgment, and system-level understanding.

Compared to traditional specialist roles, this position expands responsibility. It shifts from configuring tools to shaping and evolving agentic systems over time.

Example: A CRM expert who not only configures systems, but designs, trains, and continuously improves AI-driven lead routing, lifecycle management, and personalization agents.

Skill gap: Builds on today’s specialist roles but adds the ability to fine-tune and teach agentic systems, requiring greater technological adaptability.

3. Human Interfaces in Trust-Critical Domains

(AI-Empowered Frontline Worker)

Some parts of marketing are not defined by efficiency, but by trust:

• Sales conversations.

• Strategic partnerships.

• Negotiations.

• Leadership communication.

In these domains, human presence is not a temporary limitation. It is the value proposition.

AI agents increasingly support these interactions by providing research, insights, and recommendations. But the human remains the interface where responsibility, persuasion, and relationship-building converge.

As automation expands, weak human interaction becomes visible and costly. AI does not remove the need for human interfaces. It raises the bar for them.

Example: A key account manager who uses AI agents for account research, opportunity insights, and next-best-action recommendations, while focusing human effort on trust-building, negotiation, and long-term relationships.

Skill gap: Evolves existing frontline roles by integrating AI capabilities into jobs previously selected primarily for emotional intelligence.

Why Reskilling Is the Wrong Primary Frame

Much of the current debate frames this transformation as a reskilling challenge.

This framing is incomplete.

Most human capabilities persist even as tasks are automated. What changes is not the skill itself, but where and how it is applied.

AI fluency has increased rapidly and is now one of the fastest-growing skill categories. But AI fluency alone is insufficient.

The critical capabilities for task-driven marketing organizations are structural:

• designing systems rather than teams,

• orchestrating human–machine interaction,

• managing trade-offs under uncertainty,

• applying judgment where automation fails.

👉 The core challenge is not training individuals. It is repositioning human work inside systems.

Careers Without Stable Roles

When work is organized around tasks and workflows, careers can no longer be linear ladders tied to stable roles.

They become portfolios of positions across workflows.

Progress is defined less by title and headcount, and more by the scope of outcomes someone owns, the complexity of systems they orchestrate, and the trust they are granted.

This has far-reaching implications for

• performance management

• incentives

• leadership development

• and HR systems.

Static org charts give way to work charts that reflect dynamic ownership and accountability relationships.

Final Thought

The competitive advantage of agentic marketing organizations is not better AI. It is better design of human work. Organizations that cling to roles will optimize execution inside structures that no longer fit. They will appear efficient, yet remain fragile.

Organizations that deliberately design human work as part of their agentic architecture will scale with clarity. They will move faster without losing control. They will automate without eroding responsibility.

In the agentic era, marketing leadership is no longer about managing people.

It is about designing systems in which human judgment remains central, while speed and scale are delegated to machines.

That is the real transformation ahead.

👉 If you would like to discuss how these ideas can be applied to your marketing organization, feel free to reach out.

Yours sincerely,

Prof. Dr. Andreas Fuchs 🦊🎓

1 Sukharevsky, A., Krivkovich, A., Gast, A., Storozhev, A., Maor, D., Mahadevan, D., Hämäläinen, L., & Durth, S. (2025). The agentic organization: Contours of the next paradigm for the AI era. McKinsey & Company.

No launch campaign.

No growth hacks.

Just a simple idea: to think clearly about how marketing is changing.

Not by chasing every new tool or trend, but by slowing down enough to understand the forces beneath them.

A few months later, something quietly meaningful has emerged:

A community of CMOs, marketing leaders, practitioners, entrepreneurs, and academics who share one common trait: a genuine curiosity about what comes next in marketing … and a need for clear thinking, orientation, and actionable principles to prepare their organizations for what follows.

Thank you for that.

What I’ve appreciated most over the past months is not just the number of subscribers, but the quality of attention. Thoughtful replies. Sharp questions. Constructive disagreement. Signals that people are not here for tactics, but for understanding.

Subscribe now

The themes that emerged

Looking back at the essays published so far, a clear set of thematic pillars has emerged, structuring the different essays into coherent domains:

• Agentic Organizations & the Future of Marketing Work explore how marketing is reorganized in the AI era: the shift from roles to tasks, from managing people to orchestrating systems, and from execution to ownership in agentic environments.

• Marketing Operating Systems, Tech & Architecture focus on how marketing is built as a coherent, scalable system. These articles address operating models, tech stacks, automation, and data strategy, emphasizing architecture and integration over tool accumulation.

• AI, Search & Decision Infrastructures examine how visibility and buying decisions are changing as search turns into synthesis and customers increasingly delegate decisions to AI agents. Success moves from awareness to relevance inside machine-mediated decision systems.

• Marketing Fundamentals, Trust & the Human Signal return to first principles while confronting the limits of automation. These essays explore clarity, credibility, judgment, and responsibility … what remains uniquely human as marketing becomes programmable.

Together, these pillars reflect one core ambition: to understand how marketing is structurally changing … not just in tools, but in logic, roles, and responsibility.

The most-read articles of 2025

A few pieces clearly resonated most with readers this year:

Top #1

Top #2

Top #3

Top #4

Top #5

Top #6

Top #7

Subscribe now

Each of these essays circles around the same underlying question:

What remains human in marketing, and what must become systemic?

As we move into the new year, this question will only become more urgent. Not louder. But more structural.

For now, I simply want to say thank you for reading, thinking along, and being part of this early chapter. I’m looking forward to continuing the conversation in 2026 with even more depth, fewer buzzwords, and clearer models.

Wishing you calm days, good conversations, and a thoughtful start into the new year.

Yours,

Prof. Dr. Andreas Fuchs 🦊🎄

The integration of AI-driven workflow orchestration marks the largest organizational paradigm shift since the industrial and digital revolutions.1

What is unfolding is not another efficiency wave. It is a structural transformation of how organizations create value. Companies are evolving toward what I would call the Agentic Organization.

A system in which humans and virtual (or physical) AI agents collaborate at scale, with marginal costs approaching zero.

For marketing organizations, this shift is existential: It forces a move away from rigid hierarchies and role-based structures toward fluid, task-centered systems built around workflows, agents, automation, and orchestration.

The decisive transition is this:

From a Role-driven Organization,

manual, headcount-bound, and difficult to scale,

to a Task-driven Organization,

where value is created by orchestrated work, not by static roles.

The promise is compelling:

• Faster execution.

• Higher scalability.

• Lower marginal cost.

• Greater adaptability.

But the implications go much deeper …

Subscribe now

The Core Transformation: From Managing People to Orchestrating Work

The most fundamental impact of AI on organizational design is managerial.

Marketing is no longer about managing people. It is about orchestrating work. This reflects a broader paradigm shift across industries. Organizations move from supervising individuals to designing systems that coordinate humans and agents toward outcomes [1].

In this model, growth no longer depends primarily on adding headcount. It depends on redesigning workflows so automation accelerates value creation.

As a result, the definition of a future-proof marketing organization changes fundamentally:

• It is no longer defined by roles, titles, and reporting lines.

• It is defined by tasks and capabilities.

The central organizational question therefore shifts:

Not: “Who is responsible for what?”

But: “Which marketing tasks and capabilities exist, and how are they executed?”

This shift dismantles traditional functional silos and replaces them with flat, dynamic networks optimized for end-to-end outcomes.

The Agentic Operating Model: Workflows as the New Blueprint

In a task-driven organization, workflows become the primary unit of organizational design. Not roles. Not departments. Not headcount.

The core mistake many organizations make is to apply AI to isolated tasks inside legacy structures. This rarely delivers meaningful productivity gains. The real leverage comes from redesigning entire workflows end to end.2

Read more

Yet customers still behaved as if nothing had changed.

The next transformation will not come from marketers. It will come from customers:

• From how they decide.

• From how they search.

• From how they delegate.

AI enters the journey not as a tool, but as a proxy mind. Agents of Customers will filter reality on behalf of humans, deciding what is visible, what is credible, and what is worth considering.

With this paradigm change, the foundations of brand building and customer acquisition are not merely evolving. They are dissolving into a new logic.

Today’s Customer Journey: Heavy. Manual. Outdated.

Marketing has evolved at an extraordinary pace. Customer behavior has not. Despite a decade of technological progress, a person searching for a product still must:

• gather information

• compare alternatives

• filter noise

• assess credibility

• match choices to personal preferences.

Meanwhile, the systems around this process have grown endlessly more complex. Yet the burden of making sense of this complexity still rests entirely on the human.

Even advanced marketing organizations continue to struggle with fragmented data, inconsistent channels, and disconnected systems that impose unnecessary cognitive load on customers.

Source: McKinsey (2025).

Half of all consumers already use AI-powered search, across all age groups. And AI-powered search is now the most preferred digital information source, surpassing traditional search engines (44 percent vs. 31 percent), as shown below:

Source: Own visualization based on McKinsey (2025).

Subscribe now

The Bruce Wayne Way of Buying

Imagine you had 30 million euros in your bank account.

How would you shop?

You would not!

You would not compare washing machines. You would not decode airline pricing. You would not analyze car benchmarks at night.

Someone else would do it for you. In the world of billionaire Bruce Wayne’s, that someone is Alfred … not just a butler, but a trusted cognitive proxy. He

• absorbs complexity,

• evaluates options,

• and presents only refined conclusions.

This is the shift AI brings to everyone. It does not merely automate tasks. It automates Bruce Wayne’s butler Alfred.

AI-powered search already performs this work: synthesizing reviews, comparing features, extracting insights, and shaping recommendations before the user sees a single page.

What was once a privilege becomes a default:

In the future, every consumer becomes Bruce Wayne.

Not economically. Cognitively.

Because the real privilege is not money. It is the ability to delegate thought.

Subscribe now

The Future Customer: Assisted. Delegated. Augmented.

Consider the customer journey in the near future. Customers will no longer evaluate options manually. They will no longer crawl the internet for comparison tables. They will no longer carry the cognitive burden of decision-making themselves.

Instead, that burden shifts to an intelligent agent acting entirely in their best interest. A shopping assistant. A travel planner. A personal operating system.

• A system that understands constraints, preferences, budgets, values, and behavioral patterns.

• A system that analyzes thousands of SKUs, reviews, specifications, and price histories in milliseconds.

• A system that negotiates trade-offs and surfaces only what truly fits the individual.

And this is not a distant scenario. It is already happening:

Across major consumer categories, 40 to 55 % of buyers now rely heavily … or even exclusively … on AI-powered search as they make purchasing decisions.

In other words, the shift has already begun: Agents start curating.

At the moment this delegation becomes the norm, customers stop searching. Brinker & Riemersma (2025) describe these emerging systems as Agents of Customers, autonomous intermediaries reshaping the buyer’s journey outside the visibility of marketers.

From User Journey to Agent Journey

Today the customer journey still looks like:

trigger → search → evaluation → purchase → loyalty.

A linear model built around human effort, human cognition, and human comparison. But this sequence is already starting to dissolve. In an agent-mediated world, the journey no longer begins with a search. It begins with intent.

Soon it becomes:

The human no longer navigates the landscape of information. The agent does.

These agents increasingly operate upstream, outside brand-controlled systems, remixing information and preparing decisions long before customers ever see a webpage.

This raises a new question: Who are these agents now acting on behalf of the customer?

Subscribe now

Customer AI Agents in Marketing

The rise of Agents of Customers does not create one assistant.

It creates an ecosystem of AI agents, each shaping a different dimension of the decision-making process. These agents do not merely support the customer. They increasingly replace the customer in the upstream work of evaluation.

Here are the major categories redefining the journey:

Source: Own visualization based on Brinker & Riemersma (2025).

• AI Assistants (The Generalists): Systems like ChatGPT, Gemini, Perplexity, and Claude function as broad cognitive scaffolds. They interpret intent, refine queries, generate comparisons, extract meaning, and frame decisions. These assistants represent the first moment of delegation … the point where customers stop searching and start asking.

• Content Remix Agents (The Synthesizers): Tools such as NotebookLM, Reafwise, or Looria process the world as streams of unstructured information. They condense articles, videos, product specs, reviews, manuals, and tutorials into coherent guidance. They do the one thing no human can do effectively at scale: they metabolize the entire informational universe.

• Consumer Agents (The Decision-makers): Services like TRIM, Klarna, Hopper or Donotpay focus on specific domains of everyday life. They optimize decisions end-to-end. These agents are not informational. They are operational. They turn consumer intent into action.

• Agentic Browsers (The Autonomous Explorers): Browsers like GPT Atlas, Opera Neon, Sigma or Dia explore the web without the user present. They audit sites, scrape data, verify claims, compare prices, and return only what survives their evaluation. They do not show the internet to the customer. They show a filtered internet to the agent.

• Agentic Email (The Filters and Negotiators): Agents embedded in tools like Superhuman, Shortwave, Gmail or Canary interpret offers, prioritize messages, summarize threads, and manage outreach. They act as the gatekeepers of commercial communication. Every pitch, promotion, and brand message now passes through an agentic filter.

• Procurement Agents (the B2B forerunners): Platforms like Keelvar, Vertice, Zycus or Arkestro automate negotiation, supplier selection, and contract optimization. Historically B2B, these logics are now migrating into consumer markets.

Across these categories, a single pattern becomes clear: The customer is no longer the primary evaluator of information. Their agents are.

McKinsey quantifies the shift: 20 to 50 % of traditional organic traffic is now at risk, because decisions increasingly occur inside AI interfaces rather than on websites. The funnel collapses into the agent.

Subscribe now

Branding in the Age of AI-Mediated Customers

Brands evolve into systems of validated data, trusted signals, consistent patterns, and reliable performance … and AI agents evaluate these signals long before a human ever encounters the brand.

A brand’s meaning shifts from the consumer’s mind to the computational layer that filters reality for them. Brand equity becomes machine-interpreted equity. Perception becomes pattern recognition.

The consequences are clear: even leading brands lose visibility when their data models misalign with AI retrieval. One major sportswear brand falls to an AI visibility index of –0.6 compared to market share (McKinsey)

Storytelling still matters for humans. But schemas matter more for machines. And increasingly, the schema determines what the human will ever see.

What Marketers Must Do Today

Customer agents change the rules of marketing. The question is no longer how to persuade customers, but how to make your brand legible and recommendable to their agents.

Three priorities define this shift:

1. Design for agents and humans simultaneously: AI systems reward clarity, structure, precision, and semantic richness. If your content cannot be parsed, it cannot be recommended.

2. Build trust signals into everything: Agents privilege facts over storytelling …

   verified attributes, transparent pricing, consistent product data, validated claims. Ambiguity is a filter-out, not a friction.

3. Shift from funnels to systems: The journey is now delegated, compressed, agent-driven. Relevance emerges from the ecosystem that surrounds the customer … their assistants, browsers, email agents, and procurement logic. Yet only 16 percent of brands track their AI visibility, giving early adopters a structural advantage.

Conclusion

Customer agents change everything. They filter reality, evaluate options, and shape decisions long before a human enters the journey.

This does not adjust the old marketing model … it replaces it with a machine-first architecture of choice. Brands no longer compete for attention first, but for legibility, trust signals, and inclusion in an algorithmic shortlist.

Most organisations will keep optimising for yesterday’s customer journey. A few will build for the agent-driven one … and they will define the next decade.

And for a while, we must serve both worlds: the human journey we know, and the agentic journey emerging around it. Begin building for the agents that will soon shape every customer journey.

Long before the customer chooses you, their systems will.

Prepare for that world now.

Yours,

Prof. Dr. Andreas Fuchs 🦊

Why Data Strategy Is the First Move

Marketing has always been a game of understanding people. But in today’s world, where customer behaviour stretches across dozens of touchpoints and privacy regulations reshape data access, intuition alone collapses. What once relied on human experience now demands systemic clarity.

Modern organisations are drowning in data. Dashboards multiply. Tools expand. Storage grows. And yet the one thing that should become clearer — the customer — becomes harder to see.

This is the paradox at the centre of today’s marketing:

The reason is simple.

The problem is not the absence of data.

It is the absence of strategy.

And this matters, because data-driven marketing is no longer optional. The collapse of information asymmetry means customers now move through markets with near-perfect knowledge. They compare instantly. They evaluate independently. They expect relevance by default.

Data-driven marketing answers exactly this shift. It enables deeper …

• audience understanding

• personalised experiences

• better decisions

• more accurate journey optimisation

• improved customer relationships

• smarter product development

• and even new data-driven revenue streams.

But none of this happens automatically. None of it comes from collecting more fields or buying more tools. And none of it emerges from dashboards alone.

In every high-performing marketing organisation I work with, one pattern is unmistakable:

Data becomes valuable only when tied to a purpose … a business outcome, a decision, a process, a customer moment.

Everything else is noise.

A data strategy is not a document. It is not a dashboard. It is not a shopping list.

A data strategy is a design decision: Which customer realities must the business understand in order to grow?

Everything follows from this.

Subscribe now

Start with the Business: Why Goals Come Before Data

Before a single data point is collected, one truth must be understood:

A data strategy is only as good as the business strategy it serves.

The key to a successful data strategy lies in prioritizing business goals before data collection. Instead of starting with dashboards and tools, high-performing organizations begin with the desired outcomes.

This “Business-First” approach offers three decisive advantages:

1. Higher Data Quality: Only the targeted information that supports concrete results is collected, which reduces data waste.

2. Smarter Tech Investments: Technology is selected strategically to meet specific requirements.

3. Clearer KPIs: Important metrics (like CLV or ARR) are derived directly from business goals, rather than being incidental.

   

Source: Fuchs (2025)

The process is a clear cascade:

Source: Fuchs (2025)

Build a Use-Case-Driven Data Strategy

Once the business goals are clear, the real work begins.

And here most organisations take a wrong turn:

• They start listing data fields.

• They collect data without purpose.

• They overthink data and collect nothing at all.

But a data strategy built on lists will always collapse into abstraction. It becomes theoretical, slow, and disconnected from real impact. Data-driven marketing organisations follow a different logic:

They build their entire data strategy around use cases.

A use case is a concrete, value-creating application of data in the real world.

• It is the bridge between strategy and execution.

• It tells you exactly which data you need, why you need it, and how it creates value.

• A use-case-driven data strategy follows a clear sequence.

Here’s a practical playbook how to to it:

A Practical Playbook: How To Turn Ideas Into a Data Strategy

Once you accept that use cases are the unit of value, the next question is simple.

How do you actually build a data strategy from them?

Marr (2022)1 describes it as a structured journey from ideas to a coherent strategy. In practice, it is a five step loop that any marketing team can run, regardless of size.

Subscribe now

Step 1: Identify potentials and ideas

Read more

A promise that tools would make marketing smarter. That automation would make teams faster. That data would make decisions clearer. And that AI would soon fill the gaps humans could not.

But as the tools multiplied, something unexpected happened. The intelligence did not scale. The efficiency did not appear. And the clarity never arrived. Instead, ecosystems became bloated, budgets became strained, and organisations discovered a painful truth:

The more technology we add, the more fragile everything becomes.

Today, the average company uses only around one-third of the capabilities it pays for:

Source: Fuchs (2025) based on Gartner (2023)

The conclusion is unavoidable: Technology is not the problem. Architecture is.

Today, the real divide is not between companies with many tools and companies with few tools. It is between companies that buy technology and companies that design systems. A modern Marketing & Sales Tech Stack is not a shopping cart. It is a living, learning ecosystem that connects data, processes, and AI into a unified revenue engine.

This essay is a blueprint for how to build one.

The Real Problem: Tool Chaos Without Architecture

Modern companies live inside a labyrinth of systems.

A CRM here. A marketing automation suite there. Analytics in one place, customer service in another. E-commerce, AdTech, data warehouses, event tools, CDPs, BI dashboards … each one built to solve a small, local need.

The intention was efficiency. The outcome is fragmentation.

Every tool sees a different sliver of the customer. None see the whole.

And that creates the real crisis:

• Data scattered across channels and teams

• Customer identities split into dozens of inconsistent versions

• Marketing automating noise, not strategy

• Sales operating on half-truths

• Personalisation failing because the system cannot “see” the human behind the data

• “Real-time” existing only in product demos, not in reality

The problem is not technology overload. The problem is architectural blindness. Companies are not drowning because they have too many tools. They are drowning because the tools do not speak the same language, do not share the same memory, and do not follow the same logic.

The real question, then, is not: “Which tools should we buy?”

The real question is: “What system are we trying to build?”

Only when that becomes clear does technology finally begin to make sense.

👉 If you want the full deep dive, read my essay “The New Marketing Operating System”

Subscribe now

How a Modern Revenue Stack Actually Works

(and why most companies fail to build it)

A modern Marketing & Sales Stack is not a catalogue of tools. It is a circulatory system. Data enters, flows, connects, transforms, activates, and returns as intelligence.

Most organizations never experience this flow:

• Their data sits in silos.

• Their channels operate independently.

• Their automation triggers noise rather than intention.

To understand what a working system looks like, imagine the architecture below. It shows how data moves through a unified Revenue Engine: from raw signals to identity, from identity to action, from action to learning.

Source: Fuchs (2025)

At the top, fragmented data from warehouses, applications, and APIs is collected and harmonized.

The Cloud Data Warehouse (CDW) becomes the universal memory … the place where the entire company finally sees the same truth.

The Customer Data Platform (CDP) sits on top of this foundation. It resolves identities, builds real-time customer profiles, predicts behaviour, and prepares segments for activation. This is where “someone who clicked yesterday” becomes “someone who is ready today.”

The Orchestration Layer transforms these insights into action: Journeys fire, workflows adapt, ads synchronize with CRM updates, sales receives signals, and every digital touchpoint responds in context rather than chaos.

Finally, Engagement Channels deliver the message: ads, email, web, mobile, service, sales, apps, POS … all drawing from the same profile, all aligned around the same customer.

Every interaction flows back into Performance Analytics & Attribution, where impact is measured and the system learns what to do next.

This is not automation. This is orchestration. A self-reinforcing loop where every signal strengthens the next decision.

Composability: The Practical Secret Behind a Tech Stack

Most companies don’t fail because they chose the “wrong tools.” They fail because their tools can’t work together.

Composability solves exactly this problem.

Source: Brinker & Riemersma (2024)

Think of it like LEGO for your tech stack: You don’t buy one giant, heavy block. You build with small, modular pieces that fit together — and can be replaced, upgraded, or rearranged at any time.

Here’s what that means in practice:

• Your Cloud Data Warehouse (CDW) becomes the shared data layer for the whole company. Instead of every tool storing its own version of the truth, everything draws from one clean source.

• A Composable CDP sits on top and creates a single customer identity who they are, what they did, what they might do next. This is the engine that personalises, predicts, and activates.

• An integration layer (iPaaS) acts like plumbing. It moves data between tools, triggers actions, and keeps everything in sync.

• All your apps (marketing, sales, customer service) plug into that shared backbone. No more silos. No more guessing. No more “our numbers don’t match.”

  Subscribe now

How to Build Your Tech Stack: A Six-Step Playbook

A modern Marketing & Sales Tech Stack is not something you buy. It is something you design … deliberately, step by step.

If you ask most companies how their tech stack came into existence, the answer is always the same: not through design, but through accumulation.

• A tool bought to solve a local problem.

• A vendor pushed by IT.

• A CRM upgrade.

• A BI tool that the CFO insisted on.

• A marketing platform bought during a panic about automation.

Over years, this creates a landscape that resembles archeology more than architecture.

Source: Fuchs (2025)

The antidote is a structured, use-case-driven method for building a system that is coherent, scalable, and economically rational.

Below is the six-step blueprint:

Step 1: Start with Strategy

Read more