Every startup is interested in what influences its users. The worst way to do that is to simply ask them.

Done poorly, user research is often the worst way to understand user needs.

Oftentimes in user research, you ask people questions; Sometimes you interview them; sometimes it’s a focus group; most of the time it’s a survey; almost all of the time the information we gather is misleading.

When you ask people about past behavior, their memory fails.

When you ask people about their future, they overestimate their actions.

When you ask the golden question—”Why?”—people tend to exaggerate their rationale.

People don’t say what they think and they don’t do what they say.

They’re not lying. No, they’re human. And if there’s anything we know about humans, they’re predictably irrational.

User research is effective when you observe people’s actions and emotional states. It’s useful when you’re trying to understand existing behaviors, expectations, and, more importantly, intentions; when you’re trying to understand how your product “fits” in someone’s life.

User research is absolutely, positively counterproductive when you’re trying to generate ideas for a new feature or design.

If your product helps someone learn a new skill, build healthier habits, or be smarter with their money—you might save time and money building a strategy on principles of behavior design and skipping the questions.

There is an abundant amount of free and accessible information available on how humans make decisions.

Don’t ask questions you already know the answer to.

More importantly, don’t ask questions your users don’t know the answer to.