The person who fills in your survey is not the person who buys your product
If you’ve been here before, you may know that I have a very complicated relationship with lists. To make things even more complicated, having a system of lists (or a second brain, as some would call it) introduces an additional layer of trouble: the person who makes the list is not always the same person who uses it.
Almost every time I go to the supermarket with a list. I make it carefully, deliberately, with a clear sense of what we need for the week. I follow it, mostly. And then I come back with three things that weren’t on it and a receipt that’s twenty percent higher than it should be.
Nobody planned for the ice cream. Nobody planned for the pack of M&Ms by the checkout. Nobody planned for the extra bottle of wine that seemed like a reasonable idea at the time.
Me the planner made the list. I the doer went to the supermarket. Same person, completely different state of mind. In your business, that’s your customer research and your actual customer.
So this is not a supermarket problem. It’s a research problem.
Daniel Kahneman spent a career studying the gap between who we think we are and who we actually are when we make decisions. In his framework, there are two systems at work. System 2 is the deliberate, rational, considered version of us: the one that fills in surveys, sets intentions, and makes plans. System 1 is the fast, instinctive, emotional version: the one that actually stands in front of the shelf, clicks the button, or signs the contract.
Most customer research talks to System 2. Most purchase decisions are made by System 1.
Which means you could be interviewing the wrong person entirely, not the wrong demographic, not the wrong segment, but the wrong mental state. The person your research captures is considered, unhurried, and reflective. The person who actually buys your product is tired, distracted, and running on autopilot.
The gap between those two people is worth understanding before you build your next campaign around what customers told you they would do.
This is why focus groups are not telling the whole truth. The setting activates System 2 by design. Comfortable room, considered questions, social pressure to give thoughtful answers. Participants who are doing their best to be helpful. Nobody in a focus group says “I’d probably just grab whatever’s closest.” But that’s exactly what they do at the shelf.
Intention surveys have the same problem. “How likely are you to purchase this product in the next six months?” The answer reflects who the respondent wants to be, not who they are at the moment of decision. Stated preference is not revealed preference. And revealed preference is the only one that actually moves revenue.
This doesn’t mean focus groups and surveys are useless. They’re essential. They tell you what your customer intends to do, what they value, what they aspire to. That’s real and useful information, as long as you know what it is and what it isn’t.
When I was younger, I thought this meant everything. Ask the right questions, get the right answers, make the right decision. It took a few expensive mistakes to realise it wasn’t that simple. Life is like a box of chocolates, you never know what you’re gonna get. Oh, but that’s not what i meant.
People’s decisions are like an onion. Preference is just one layer. To get closer to the real decision, the one made under pressure, in a hurry, with incomplete information, you need to peel a little further.
So how do you do it?
Behavioural data is the most honest research you have. What people actually did (clicked, bought, abandoned, returned) is System 1 speaking without a filter. It doesn’t tell you why, but it tells you what. Start there.
Observational research captures behaviour as it happens: watching customers in context, not asking them to reflect on it afterwards. A customer navigating your app while slightly distracted is a different creature from one describing how they’d navigate it in an interview.
Jobs to Be Done interviews don’t ask about preferences, they ask about episodes. “Walk me through the last time you actually did this.” Specific, past-tense, grounded in a real moment. Much harder to idealise than a hypothetical, and much closer to System 1 than any focus group.
Context and friction gets you the why behind the what. Where do customers drop off, work around the product, or do something you didn’t design for? Friction is System 1 speaking out loud. Behavioural data shows you the symptom. Context shows you the cause.
None of this requires a permanent research budget. Most of it is already sitting in your CRM, your analytics, your support tickets, your lost deals. System 1 has been speaking all along. You just have to listen to the right channel.
When the stakes are higher, a new launch, a drop in sales, a market you don’t fully understand, that’s when you bring in a market research company. Use them to measure intention and map preferences. Then test what they find against what your customers actually do. The gap between the two is not a research failure. It’s the most useful thing you’ll learn all year.
The planner and the doer are both your customer. Research introduces you to one of them. The other one you have to go find yourself.
Further Reading
- Daniel Kahneman — Thinking, Fast and Slow
- Clayton Christensen — Competing Against Luck
- Phil Barden — Decoded: The Science Behind Why We Buy
Leave a comment