Why Some UX Insights Should Never Become Metrics

By Philip Burgess | UX Research Leader

When I first started working in user experience, I was eager to measure everything. I believed that turning every insight into a metric would help prove the value of UX work and guide product decisions. But over time, I learned that not all UX insights fit neatly into numbers. Some insights lose their meaning or even mislead when forced into metrics. This post shares why some UX insights should stay qualitative and how to decide what to measure.

Designer sketching user flow on paper

The Limits of Metrics in UX

Metrics are powerful. They provide clear, comparable data that can track progress and highlight problems. But UX is about human behavior, emotions, and context—things that don’t always translate well into numbers.

For example, a user’s frustration with a confusing interface might show up as a high bounce rate or longer task time. But those numbers don’t explain why users feel frustrated or what exactly causes it. If you only look at metrics, you might miss the root cause or fix the wrong problem.

I once worked on a project where the team focused heavily on reducing task completion time. The metric improved, but user satisfaction dropped. We realized that users rushed through tasks because the interface was unclear, not because it was efficient. This insight came from interviews and observation, not metrics.

When UX Insights Should Stay Qualitative

Some insights are best captured through stories, quotes, or observations. These include:

• Emotional reactions: How users feel during interactions can reveal pain points or delights that numbers can’t capture.

• Contextual behaviors: Why users behave a certain way often depends on context, environment, or mindset.

• Unexpected discoveries: Open-ended research can uncover needs or problems you didn’t anticipate.

  
  

Turning these into metrics risks oversimplifying or missing nuances. For example, measuring “user happiness” with a single survey question can’t capture the complexity of emotions users experience.

How to Decide What to Measure

Not every insight should become a metric, but some do need measurement to track changes or validate hypotheses. Here’s how I decide:

• Is the insight clear and specific? Vague feelings or broad observations don’t make good metrics.

• Can it be measured reliably? If the data is inconsistent or hard to collect, the metric won’t be useful.

• Will the metric guide action? Metrics should help teams make decisions, not just report numbers.

• Does it complement qualitative insights? Metrics work best alongside stories and observations, not as a replacement.

  
  

For example, if users say a checkout process feels slow, measuring the actual time to complete checkout makes sense. But if users say they feel “confused,” that insight needs more exploration before creating a metric.

Balancing Metrics and Qualitative Insights

The best UX research combines both. Metrics show trends and patterns, while qualitative insights explain the reasons behind them. I recommend:

• Use metrics to track progress on clear goals like task success rates or error counts.

• Use qualitative methods like interviews, usability tests, and diary studies to explore user feelings and context.

• Share stories and quotes with stakeholders to bring data to life.

• Avoid forcing every insight into a number just because it seems easier to report.

  
  

UX researcher observing user interacting with mobile app

Real-World Example: Improving a Mobile App

In one project, users reported feeling overwhelmed by too many options on the home screen. We could have tried to measure “overwhelm” with a survey question, but that wouldn’t tell us what overwhelmed them or how to fix it.

Instead, we conducted usability tests and observed users struggling to find key features. We used those insights to simplify the layout and prioritize important actions. After the redesign, task success rates improved by 20%, and users reported feeling more confident. The metric tracked the outcome, but the insight came from qualitative research.

Final Thoughts

Not every UX insight should become a metric. Some insights lose their meaning when reduced to numbers, while others need more context to guide action. Metrics are useful tools, but they don’t replace the rich understanding that qualitative research provides.