Part 1 of Marketing in the Age of AI. The series overview is at /insights/articles/marketing-in-the-age-of-ai/.
Ask most people what marketing is and you will hear some version of advertising, or social media, or the team that makes the campaign. That answer describes one slice of the work. Marketing is the discipline of understanding demand and organising a business to meet it profitably. Advertising is a visible box inside it, but it is only one box.
That distinction matters now, because the wave of tools sold as "AI for marketing" is landing almost entirely in that one box. If you want to know where AI changes the economics of a marketing function, you first need the whole chain in front of you.
The chain from end to end
Marketing rests on three questions asked in sequence.
- WHO is the customer, and what do they need?
- WHAT proposition do we build for them?
- HOW do we deliver it and capture the value?
Everything a marketing function does hangs off those three questions. Broken into the work itself, it looks like five linked pieces:
| # | Piece | What lives here |
|---|---|---|
| 1 | Understand | Demand, ICP and segmentation, jobs-to-be-done, buying triggers, insight |
| 2 | Strategise | Where to play and how to win, positioning, brand purpose, portfolio |
| 3 | Create | Product, proposition, innovation, pricing |
| 4 | Communicate | Content, creative, channels, media, campaign |
| 5 | Measure | Brand equity, marketing-mix modelling, attribution, brand value |
Read left to right, it is a value chain. You understand a market, you decide where to compete, you build something worth buying, you tell people about it, and you measure whether any of it moved the business. Skip a box and the chain leaks: brilliant communication of a weak proposition still sells a weak proposition.
The concentration of AI tools
Overlay the current generation of AI marketing tools on that chain.
The tools generate the post, write the email, draft the ad, produce ten headline variants and spin up a landing page. Almost all of it lives in box four, Communicate. That is the loud, visible, easy slice of marketing, and it is where the demos look most impressive, because producing a plausible paragraph or a passable image is exactly what a language model does out of the box.
There is nothing wrong with that. Box four is real work, and doing it faster has value. The problem starts when a team that has bought AI tools for marketing concludes that AI is improving its marketing, while the tools only ever touched one box.
Box four is also the slice where human judgement was already cheapest. A skilled copywriter turning a clear brief into a good ad was never the bottleneck. The bottleneck was upstream, in working out who to talk to and what to say, and downstream, in working out whether it landed.
The leverage in the chain
The leverage in a marketing function, and the hard judgement, sits in boxes one, two, and five: Understand, Strategise and Measure. Automate only box four and you scale output without improving any of the decisions that make output worth producing. You get more posts about a proposition that may be aimed at the wrong buyer, measured by a dashboard nobody trusts.
The teams pulling ahead with AI are the ones pointing it at the expensive boxes:
- Understand is where research used to be slow and infrequent. A segmentation study runs into six figures and takes three to four months, so it happens rarely and the insight is often stale by the time it reaches a brief. That is the box where AI changes the unit economics most, and it is the one the tool market has largely ignored.
- Measure is where most functions quietly give up. Attribution is hard, brand equity moves slowly, and few teams can say whether a given campaign built the brand. It is a box built for systems that read a lot of signal and hold a consistent model over time.
Boxes one and five are covered in their own right later in this series - the collapsing cost of research and the shift to a live measurement function are each a piece on their own. The point here is structural: the value chain tells you where to look, and the market is looking in the wrong place.
The discipline is a system
This matters because each box has rigour underneath it. The discipline is a system you can reason about, so AI can be aimed at each box on purpose.
Consider box one. Growth has only three levers. You win new customers, you expand the ones you have, or you retain them: land, expand, retain. Every B2B go-to-market plan, stripped down, reduces to those three, and the job is deciding which one you are playing this year. That is a decision you can bring evidence and models to.
Consider box two. In B2B, becoming a client's preferred partner has a measurable structure. Kantar's work on B2B relationships finds that the strength of the client relationship is the single largest driver of preference - on the order of a quarter of the outcome - ahead of client experience and distinctiveness, and well ahead of awareness or a clever tagline. Positioning, in other words, has a shape you can test against, and part 3 of this series examines that model in full at /insights/articles/b2b-preferred-partner-positioning/.
None of this depends on any one framework being the final word. The frameworks come from consulting and training bodies of work: the Metro Marketing Academy material and the Kantar Vermeer toolkits among them. The shape of the thing holds regardless of the source: marketing is a chain of linked, analysable decisions, and AI touches every box of it once you stop treating it as a copy machine.
If your AI plan for marketing currently lives entirely in box four, it fits the pattern this article describes. The next three parts of this series cover the boxes where the leverage sits: what changes before and after AI, how positioning holds up as a model, and what a live measurement function looks like when research stops being a once-a-year event.
That is the lens we bring to a market before we write a word of positioning. We built NavigatorLab to apply it to a real category question: where growth lives in your market.
Sources and further reading
- Kantar, B2B relationship and preferred-partner research - kantar.com
- Metro Marketing Academy and Kantar Vermeer marketing toolkits (consulting and training material)