9 Cognitive Biases of a Product Manager: How Not to Deceive Yourself During Validation
An overview of common cognitive biases (Confirmation Bias, Polite Lies, Sunk Cost, and others) that prevent product managers from getting honest signals during validation and how to combat them.
During the idea validation stage, product managers often fall victim to their own thinking. Our brains are wired to seek shortcuts and confirm our existing beliefs, which can lead to the creation of "eternal clutter"—features that no one needs.
Knowing these traps is the first step to avoiding them. Here are 9 common cognitive biases that lie in wait for product managers, and ways to combat them.
1. Confirmation Bias
We subconsciously seek, interpret, and remember information that confirms our pre-existing beliefs.
- How it looks for a product manager: After conducting 10 interviews, they focus on two responses that support their idea, declaring the other eight "irrelevant" or "not the target audience."
- Antidote: Define
kill-criteriain advance—clear conditions under which a hypothesis will be considered disproven. For example: "If less than 20% of users in the test do not perform the target action, we scrap this idea."
2. Social Desirability Bias (Polite Lies)
People tend to give answers that they believe will be perceived positively, even if they are not true.
- How it looks for a product manager: To the question "Would you use this?", almost everyone answers "Yes, of course!". But when you ask them to leave an email for early access or click a fake-door button, no one does.
- Antidote: Ask questions about the past and facts, not the future and opinions. Instead of "Would you...?", ask "Tell me, how did you solve this problem last time?" (This is the main principle of "The Mom Test").
3. Selection Bias
We draw conclusions based on a group of people that is not representative of our entire market.
- How it looks for a product manager: Interviews are conducted only with loyal "power users." As a result, complex "turbo-features" are created that only they need, while the mass market still cannot grasp the basics.
- Antidote: Set sampling quotas and strive for diversity. Include beginners, churned users, and those who use the product differently than you expected in your research.
4. Authority Bias
We tend to overvalue the opinion of an authoritative figure and follow it, even if it contradicts facts.
- How it looks for a product manager: "The CEO saw this at a competitor's—let's do it immediately." The validation process turns into a formal search for arguments "for."
- Antidote: Any statement from an authority, whether a CEO or a well-known expert, should be treated as a hypothesis. It must undergo the same strict vetting process as any other.
5. Anchoring
We rely too heavily on the first piece of information received when making decisions.
- How it looks for a product manager: At the start of a project, someone hypothesized that "conversion will be 20%." When test results show 7%, the team doesn't abandon the idea but explains it as "well, almost, we just need to push harder."
- Antidote: Instead of a single "anchor" figure, define a range of expectations and a clear failure threshold. For example: "Below 5%—stop."
6. Sunk Cost Fallacy
We continue to invest in a project because we have already put a lot of resources (time, money) into it, even if it is clearly failing.
- How it looks for a product manager: "We've already designed 40 screens and spent a month discussing. We can't just throw it all away now, let's build it."
- Antidote: Make decisions based solely on future cost and future benefit. Past expenses are sunk investments; they cannot be recovered.
7. Halo Effect
Our overall impression of something (a person, a brand) influences our evaluation of its specific characteristics.
- How it looks for a product manager: One VIP client says "I'll take it!". The team ignores the fact that this client has unique processes, a huge budget, and a non-standard context, and decides that the feature is needed by everyone.
- Antidote: Conduct a "transferability" check. Ask yourself: "What must be the same in other clients' processes for this VIP client's experience to be relevant to them?"
8. Optimism Bias & Planning Fallacy
We tend to overestimate the likelihood of positive outcomes and underestimate the time required to complete a task.
- How it looks for a product manager: "We'll build an MVP in two weeks!" Then it turns out that complex integrations, support training, legal review, and data migration are needed.
- Antidote: Use the premortem technique. Imagine that two months have passed, and your project has completely failed. What went wrong? Turn these causes of failure into risks that need to be checked at the earliest stage.
9. Goodhart's Law
When a measure becomes a target, it ceases to be a good measure.
- How it looks for a product manager: The goal is to "increase activation." The team adds aggressive modal windows and mandatory steps. The metric "clicks on onboarding steps" grows, but actual activation and retention fall because users are simply annoyed.
- Antidote: Always use guardrail metrics. Ask yourself: "How can we improve this number while simultaneously worsening the product?" This will help identify potential negative consequences.
By recognizing these traps, you can approach the validation process more objectively, making decisions based on honest signals, not cognitive illusions.