System 1 and System 2 thinking in marketing: what the two-system model says, and what marketers added to it

Origin

The idea that thought runs on two tracks is old. The specific vocabulary is recent. Keith Stanovich and Richard West proposed the terms System 1 and System 2 in a 2000 paper in Behavioral and Brain Sciences, summarising a family of dual-process theories that had been developing in cognitive psychology since the 1970s [2]. They used the numbers deliberately, to avoid committing to any one theory’s descriptive labels.

Kahneman adopted the terms for Thinking, Fast and Slow in 2011 [1]. The book was a summary of four decades of work with Amos Tversky and others on judgement and decision making, and the two systems were the frame that organised it. System 1 produces impressions, intuitions and feelings quickly and without effort. System 2 allocates attention to effortful operations, including the ones we call reasoning. Most of the time System 2 accepts what System 1 offers.

Kahneman was careful about the status of the terms. He wrote that the two systems are “fictitious characters” [1], useful because the mind finds it easier to follow a story about agents than a description of processes. The caveat is in the book. It did not make it into the marketing version.

The problem the model solved in psychology was explanatory. It gave a common language to findings about heuristics and biases that had been accumulating without one.

What the evidence shows

The underlying findings have held up. The specific effects that Thinking, Fast and Slow describes, anchoring, availability, loss aversion, framing, the planning fallacy, have decades of replication behind most of them. They are the content. The two systems are the frame put around the content. One can accept the findings and treat the frame as the convenience Kahneman said it was.

The frame is an organising device, not a discovery. This is the point most often lost. No experiment found System 1. The term labels a cluster of properties, fast, automatic, associative, that tend to go together, and Stanovich himself later moved away from the simple two-system language toward a more layered model [3]. A marketer presenting the two systems as two things the brain contains has gone further than the people who named them.

The critiques have teeth. Melnikoff and Bargh argued in 2018 that the properties assigned to each system do not in fact align: automatic processes can be slow, deliberate ones can be fast, and the typology breaks down under inspection [4]. Keren and Schul had made a related case in 2009 [5]. Neither critique disputes the individual findings. Both dispute the tidy division.

What marketing can legitimately take. Most purchasing is habitual and low attention. This is well supported from the buying behaviour literature independently of Kahneman [6]. If the fast mode handles most choices, then the work of a brand is to be present and easy at the moment of choice: recognised at a glance, associated with the category, requiring no deliberation. That is a modest claim and a well-founded one. It is also, in substance, what Byron Sharp’s work says in different vocabulary.

The challenge

The challenge to the two-system model inside psychology is about the frame, not the findings, and this entry has set it out. The challenge to the marketing use is different and more pointed.

Marketing took a model of how judgement works and turned it into a model of how advertising should work. The inference was: if System 1 decides, make advertising for System 1. Emotional, implicit, fast. Rational argument is for System 2, and System 2 is asleep.

Three things are wrong with this. First, it is not in the book. Kahneman wrote about judgement under uncertainty, not about advertising. Second, it collapses the distinction between how a choice is made and what information went into making it. A habitual choice can rest on a rational judgement made once, years ago, which is how most people choose their bank. Third, it has produced a genre of marketing that treats the audience as incapable of thought, which the audience notices.

The two-system language has also been used to dress up a much older argument, that emotional advertising outperforms rational advertising, with a neuroscience it does not have. That argument has its own evidence, from campaign effectiveness data, and does not need to borrow.

Where I see this go wrong

System 1 becomes a licence. I have sat in presentations where “it’s a System 1 decision” was offered as the reason to drop every claim from the work. The reasoning went: buyers do not think, so give them nothing to think about. In a B2B category where the buyer is a committee with a procurement process, this is not insight, it is an excuse.

Two systems become two audiences. The model describes two modes in one mind. It gets applied as if some buyers were System 1 people and others System 2 people, with different messaging for each. There is no such division. Every buyer is both, at different moments, and the same person who recognises your brand in a second will also read your tender response line by line.

The fictitious characters become anatomy. Slides that locate System 1 in one part of the brain and System 2 in another are common and wrong. Kahneman said the systems were a story. The slide has turned the story into a diagram of an organ.

Having spent years in television before this, I recognise the move. A broadcaster who wants to defend low-effort programming will tell you the audience is tired and wants nothing demanding. Sometimes true. Often an alibi. System 1 is the same alibi with a Nobel laureate’s name on it.

What to do with it

1. Use the findings, hold the frame loosely. Anchoring, framing and loss aversion are tools you can test. The two systems are a way of talking about them.
2. Accept the modest claim: most choices in your category are fast and habitual, so your brand must be recognisable and easy to pick. Build for that. Distinctive assets, consistent presence, nothing in the way.
3. Reject the grand claim. “Bypass reason” is not a strategy. Your buyer will reason at some point in the journey, and the work must survive that moment too.
4. In B2B especially, map where the fast moments and the slow moments actually fall. Recognition on a shortlist is fast. Reading the proposal is slow. Different work serves each. Both are needed.
5. Never put a brain diagram on a slide about the two systems. If challenged, quote Kahneman on fictitious characters and move on.
6. Read Stanovich and West before Kahneman if you want the model, and read Melnikoff and Bargh after. The book is the popular account. The paper is the source. The critique is the correction.
7. When someone says “this is a System 1 category”, ask what they would do differently if it were not. If the answer is nothing, the label is doing no work.

If you run the company rather than the marketing

You will hear that your buyers make System 1 decisions, and it will be offered as a reason to spend on brand rather than on argument.

Half of that is sound. Most choices, including expensive ones, are made quickly from a short mental list, and getting onto that list is what brand spending does. The other half, that argument therefore does not matter, is not in the research and is not true of any buyer who signs a contract. The committee that approves your supplier will read the proposal. The person who champions you will have to defend the choice. Both are slow, effortful moments and the work has to hold up in them.

The question to ask is where the fast and the slow moments fall in your own sales process, and whether the plan serves both. If the agency can only describe one, you are getting half a strategy.