Confirmation Bias: Your Customers Already Made Up Their Minds

A 1960 experiment that still explains your buyers

In 1960, Peter Wason ran a deceptively simple experiment at University College London. He gave participants three numbers, 2, 4, 6, and told them the sequence followed a rule. Their job: figure out the rule by proposing their own sequences. Wason would tell them yes or no.

The actual rule was almost insultingly simple: any three ascending numbers. 1, 2, 3 works. So does 5, 19, 400.

But participants who saw 2, 4, 6 almost universally assumed the rule was "ascending even numbers" or "numbers increasing by two." So they tested sequences like 8, 10, 12 or 20, 22, 24, which all fit Wason's actual rule. He said yes. They grew more confident. And they never tried something like 1, 3, 7, which would have also gotten a yes and blown their theory apart.

Around 80% of participants — 23 out of 29 — announced an incorrect rule at least once before getting it right. Many got it wrong multiple times. They weren't stupid. They were doing what humans do: searching for evidence that confirmed what they already believed, and ignoring anything that might prove them wrong.

Wason called this confirmation bias. It became one of the most replicated findings in cognitive psychology.

This isn't a bug in reasoning. It's how reasoning works.

Raymond Nickerson reviewed confirmation bias across dozens of studies in a 1998 paper for Review of General Psychology. He opened by citing Evans (1989), who called confirmation bias "perhaps the best known and most widely accepted notion of inferential error to come out of the literature on human reasoning."

He found it everywhere. Medical diagnoses, courtroom decisions, scientific research, everyday judgment. It showed up regardless of education level, domain expertise, or cognitive ability. Smart people aren't immune to confirmation bias. They're often just better at building arguments for what they already believe.

Klayman and Ha refined the picture in 1987 with what they called the "positive test strategy." Their argument: what looks like confirmation bias is often something subtler. People tend to test hypotheses by looking for cases where the expected property is present, rather than looking for cases where it should be absent. In a lot of real-world situations, that's actually a reasonable shortcut. But when someone's deciding whether to buy something, where the stakes feel personal and the options seem endless, it leads to a predictable pattern.

People don't research to discover the best option. They research to confirm the option they're already leaning toward.

How buyers actually use information

Think about the last big purchase you made. Did you genuinely weigh every option equally? Or did you have a front-runner within the first few minutes and then spend the rest of your research looking for reasons that front-runner was right?

If you're honest, it was the second one.

A 2024 review by Honka, Seiler, and Ursu in the Journal of Retailing examined what pre-purchase search data reveals about consumer behavior. The findings lined up with what behavioral economists have long suspected: consumers spend most of their research time on options they already prefer, and they seek out information that validates that initial preference.

Algorithmic filtering makes it worse. Google, Amazon, and social media all serve personalized content based on what the user has already shown interest in. The consumer thinks they're doing independent research. They're walking a confirmation loop.

So what does this mean for testimonials? A testimonial doesn't change someone's mind. It gives them permission to act on a decision they've already made.

Testimonials do different things at different stages

Most marketing advice treats testimonials as one tool with one job: convince people to buy. But confirmation bias means testimonials do completely different work depending on where the prospect already stands.

People who are problem-aware know something's wrong but haven't identified a solution. They're googling things like "why is my conversion rate low" or "how to get more customer feedback." For them, testimonials that describe the problem, not your product, are what land. A testimonial saying "I was spending 20 hours a month chasing customer reviews manually" confirms their suspicion that the problem is real and solvable. That nudge moves them forward.

People who are solution-aware know they need a certain kind of tool but haven't picked one yet. They're comparing. Testimonials that describe the experience of switching, the before and after, confirm that this category of solution actually works. "We tried collecting testimonials through email and it was chaos. Having a dedicated tool changed everything." That confirms their growing belief that they need a specialized tool, not another workaround.

People who are product-aware know your product exists and are deciding whether to buy it. Now the testimonials need to confirm that your product specifically is the right pick. Concrete features, concrete results, concrete use cases. "PraiseLane's embeddable widget took 10 minutes to set up and our testimonial page conversion went up 34% in the first month." That confirms their lean toward you.

Get the stage wrong and you don't just waste the testimonial. You might trigger the bias in reverse. A product-specific testimonial shown to someone who's still problem-aware feels like a sales pitch. And once they categorize you as "trying to sell me something," they start looking for reasons to dismiss you.

More testimonials won't save a bad first impression

This is the part nobody wants to hear. If a prospect already has a negative impression, whether from a confusing landing page, a clunky first interaction, or a critical review they saw first, piling on more testimonials won't fix it. Confirmation bias is now working against you.

Nickerson documented this in detail. Once someone forms an initial hypothesis, they interpret ambiguous evidence as supporting it. A prospect who already thinks you're overpriced reads a testimonial that mentions pricing and thinks "see, people have to justify the cost." A prospect who thinks you're the right choice reads the same testimonial and thinks "see, people think it's worth it." (This is also why imperfect reviews often outperform flawless ones -- they feel more authentic to prospects who are already leaning favorably.)

Same testimonial. Same words. Opposite conclusions.

This is why first impressions, onboarding, and initial touchpoints matter more than social proof volume. You can't wallpaper over a bad first impression with positive reviews. The prospect's brain will find ways to discount, reinterpret, or just skip them.

What you can do is figure out where first impressions actually form: your homepage headline, your ad copy, your pricing page layout, that first email. Make those moments create a positive initial lean. Once someone leans your way, confirmation bias flips from obstacle to advantage. Every testimonial, case study, and data point just reinforces the decision they already want to make.

Segment your testimonials by awareness stage

The practical takeaway is simple, even if most companies skip it. Stop dumping all your testimonials into one undifferentiated pool. Segment by awareness stage and serve them in context.

At the top of funnel (blog posts, educational content, problem-focused landing pages), use testimonials that describe the problem your prospects face. These confirm the prospect's feeling that the problem is real and worth solving.

In the middle of funnel (comparison pages, feature tours, case studies), use testimonials that describe the transformation. Before and after. The path from problem to solution. These confirm the category works.

At the bottom of funnel (pricing pages, signup flows, trial pages), use testimonials with specific results, features, and use cases. Names, numbers, timelines. These confirm that this product is the right one. (Getting that specificity requires asking the right questions in the first place.)

With PraiseLane, you can tag and categorize testimonials as you collect them, by customer stage, use case, and the specific outcomes they mention, then embed the right ones on the right pages. It's not about having more social proof. It's about having the right proof at the right moment, when it'll actually confirm what the prospect is starting to believe.

Companies that get this don't try to change minds. They figure out which way minds are already leaning and make it easy to follow through. Testimonials aren't persuasion tools. They're confirmation tools. That distinction is the gap between marketing that converts and marketing that gets scrolled past.

What this actually means for your strategy

Wason's participants in 1960 weren't trying to be irrational. They were being efficient. Testing confirming evidence feels productive because you get positive feedback. Testing disconfirming evidence feels risky because it might destroy a perfectly good hypothesis. Our brains are wired to conserve energy, and restarting an evaluation from scratch is expensive.

Your prospects do the same thing. They show up to your marketing materials with a lean, sometimes toward you, sometimes away, and they're looking for evidence to justify acting on it.

You're not going to flip that lean in a single interaction. That almost never works. What you can do is figure out what created the lean, make sure as many people as possible arrive with a favorable one, and then deliver the right confirmation at the right time.

Wason gave us the science. Nickerson gave us the review. Klayman and Ha gave us the nuance. They all point to the same thing: people don't seek truth when they're buying. They seek validation. Build your testimonial strategy around that, and you'll stop fighting human nature and start working with it.

---

Sources:

• Wason, P. C. (1960). "On the Failure to Eliminate Hypotheses in a Conceptual Task." Quarterly Journal of Experimental Psychology, 12(3), 129-140.
• Nickerson, R. S. (1998). "Confirmation Bias: A Ubiquitous Phenomenon in Many Guises." Review of General Psychology, 2(2), 175-220.
• Klayman, J. & Ha, Y.-W. (1987). "Confirmation, Disconfirmation, and Information in Hypothesis Testing." Psychological Review, 94(2), 211-228.
• Evans, J. S. B. T. (2016). "Reasoning, Biases and Dual Processes: The Lasting Impact of Wason (1960)." Quarterly Journal of Experimental Psychology, 69(10), 2076-2092.
• Honka, E., Seiler, S. & Ursu, R. (2024). "Consumer Search: What Can We Learn from Pre-Purchase Data?" Journal of Retailing, 100(1), 114-129.
• Schwarz, N. (2004). "Metacognitive Experiences in Consumer Judgment and Decision Making." Journal of Consumer Psychology, 14(4), 332-348.