Multidisciplinary marketing in the age of AI

I have never been very good at staying inside one subject.

I can begin with a Google Ads campaign and end up reading about attention, reinforcement learning, memory, uncertainty, behavioral economics, or the architecture of the internet carrying the ad to the screen. For a long time I treated this as intellectual restlessness. Useful sometimes, distracting often.

I now think it describes the kind of marketer the age of AI will reward.

AI can already write twenty headlines, summarize a customer interview, inspect a spreadsheet, produce creative variations, draft a landing page, and suggest a campaign structure. These abilities will improve. The production of plausible marketing will become abundant.

But abundance does not tell us which problem is worth solving. It does not tell us whether the customer is afraid, confused, indifferent, constrained by price, unable to trust the claim, or simply looking at the wrong offer at the wrong moment. It does not tell us which evidence should change our mind.

The scarce skill moves upstream: seeing.

The brain is not a buying button

Neuroscience attracts marketers for an understandable reason. Marketing is about attention, memory, motivation, learning, emotion, and choice. Neuroscience studies the machinery that makes these possible. The connection seems irresistible.

It is also very easy to abuse.

Put an image of a brain beside an ordinary persuasion principle and it suddenly appears more scientific. Mention dopamine and almost any behavior can be made to sound explained. Say that the amygdala responded, or that a color activated the reward system, and a weak claim acquires the costume of mechanism.

But a mechanism is not a magic word. Dopamine is not desire in liquid form. Emotion is not the opposite of reason. There is no universal circuit for making somebody click "buy now."

It may prompt us to ask whether the message will be noticed at all. Whether working memory is overloaded. Whether the customer can connect the claim to a concrete future self. Whether repetition is building familiarity or merely irritation. Whether the reward is immediate while the cost is abstract, or the cost immediate while the reward is distant.

These are better questions. They still need psychology, customer research, and behavioral evidence before they become marketing decisions.

People do not encounter an ad as isolated brains

A person sees an ad inside a life.

They may be tired, hurried, ashamed, hopeful, skeptical, lonely, experienced in the category, or completely unfamiliar with it. They carry habits, identities, relationships, budgets, cultural associations, and memories of every company that made a similar promise and failed to keep it.

Psychology helps us think about perception, avoidance, social proof, uncertainty, goals, identity, and the difference between what people say and what they do. Economics reminds us that preferences live inside constraints. Anthropology and sociology remind us that meaning is shared, not generated by an individual nervous system alone. Statistics protects us from turning a vivid story into a universal law. Design determines whether the idea becomes perceptible. Technology determines whether the experience works.

Marketing lives in the connections between these fields.

Consider a landing page with a low conversion rate. One discipline gives us one kind of diagnosis:

• Advertising asks whether the traffic matches the offer.
• Psychology asks whether the promise is credible and the next step feels safe.
• Cognitive science asks whether the page demands too much attention and working memory.
• Economics asks whether the price, risk, and alternatives make action rational.
• Design asks whether the hierarchy makes the intended action obvious.
• Analytics asks whether the apparent problem is even real.
• Engineering asks whether the page is slow, broken, or measured incorrectly.

None of these explanations should win because it sounds most sophisticated. They become hypotheses. We look for evidence, make a bounded change, and observe what happens.

Attention is not enough

Marketers talk about attention because without it nothing else can happen. But attention alone is a poor objective. A car crash gets attention. Confusion gets attention. Outrage gets attention.

The more useful sequence is something like this: attention must become interpretation; interpretation must attach to memory; memory must remain available when a decision becomes possible; and the decision must survive the friction of reality.

This is where one of my recurring interests, narrative continuity, enters marketing. Information becomes more powerful when a person can connect it to a story about themselves across time:

This explains my present problem. I can imagine the future in which it is solved. This action is the bridge between them.

A feature list often fails because it remains semantically understandable but personally inert. The customer can repeat what the product does without seeing where it belongs in their life. Good marketing does not merely transmit facts. It helps relevant facts become action-guiding.

That does not mean manufacturing insecurity or manipulating someone into a choice against their interests. The best marketing reduces the distance between a real problem and a useful decision. If the product cannot carry the promise, persuasion only accelerates disappointment.

AI makes synthesis more valuable

Before AI, multidisciplinary work was expensive partly because gathering the material was expensive. Reading across fields takes time. Translating concepts between them takes more. Most professional incentives reward specialization, and most deadlines reward whatever explanation is already familiar.

AI changes the cost of exploration. A marketer can interrogate a statistical result, compare psychological models, inspect implementation details, summarize research, and prototype an experiment in the same afternoon.

This is a remarkable extension of curiosity. It is not a replacement for it.

The danger is that AI also makes shallow interdisciplinarity cheap. It can produce an impressive paragraph connecting oxytocin, color psychology, loss aversion, and conversion rates whether the connection is justified or not. Fluency can hide the point where evidence ended and association began.

So the multidisciplinary marketer needs two opposing capacities:

1. The imagination to form connections across domains.
2. The discipline to break those connections when reality does not support them.

AI helps generate models. The human remains responsible for choosing the smallest model sufficient for the next decision, marking uncertainty, and designing the contact with reality that might disprove it.

The marketer as investigator

I increasingly think the best model for a marketer is not a persuader but an investigator.

The investigator begins with an anomaly. Why are people clicking but not buying? Why does the message work for one audience and fail for another? Why did performance change after an intervention? Why do customers praise a feature they rarely use?

Then comes a rhythm:

1. Observe the behavior without rushing to explain it.
2. Ask what different disciplines make visible.
3. Separate what is known from what is plausible.
4. Choose a hypothesis that can alter a real decision.
5. Run the smallest useful experiment.
6. Keep the result and the reasoning, not every thought produced along the way.

This is also the kind of relationship I want between people and AI inside Vataman. Not an oracle issuing confident recommendations. A system that can gather perspectives, reconstruct context, expose assumptions, preserve evidence, and help the operator decide what to test next.

The future of marketing will contain more machines, more data, more generated creative, and more automated decisions. That makes the human capacity to connect biology with biography, numbers with motives, and mechanisms with meaning more important.

The point of learning across disciplines is not to make marketing sound intellectual. It is to become less wrong about people.