Evidence Does Not Speak for Itself

FIELD NOTE 004 · INTERPRETATION

Why the same research can lead reasonable people to very different conclusions

I have been in research sessions where the consumers seemed remarkably clear. They described what they were experiencing, what frustrated them, what they had tried, and what they wanted instead. We listened to their words, watched what they did, and noted the distance between the experience they had and the one they wanted.

Then we returned to the organization. One person remembered a particular quote. Another focused on the numbers. Someone viewed the findings through the products the company already offered. A technical colleague saw risk. A business colleague saw opportunity. Someone else questioned whether the evidence was strong enough to act on.

The research had not changed. Our interpretations of it had. That distinction matters because we often talk about research as though it moves cleanly from evidence to insight to action. In reality, there are people at every step, and people bring their own experience, responsibilities, priorities, assumptions, and incentives to what they hear. The same finding can therefore produce several different conclusions without anyone necessarily being careless or unreasonable.

A technical team may be listening for feasibility. A business team may be thinking about cost or market potential. A senior leader may be comparing the opportunity with other priorities. The researcher may still be thinking about the context surrounding what the participant actually said.

All of those perspectives can add value. The risk comes when we stop recognizing that interpretation is taking place. One of the most useful distinctions I have learned to make is between what we observed and what we think it means.

Imagine watching someone repeatedly move items around while trying to complete a task. One person might conclude that she needs more storage. Another might believe the layout is confusing. Someone else might see a need for faster access or a different way of organizing the space. The observation has not changed. The interpretation has. And each interpretation points toward a different solution.

Problems arise when the interpretation becomes so tightly attached to the observation that the two begin to feel like the same thing. Instead of returning to what we actually saw and heard, teams can quickly find themselves debating whose conclusion is right. This becomes especially important when the people making the decision were not present for the research. A lived experience becomes a summary. The summary becomes a presentation. The presentation becomes a recommendation. Eventually, the recommendation becomes a decision.

At every step, meaning can become clearer. It can also become more distant from its source. This is one reason I am cautious when the immediate response to disagreement is, “We need more research.” Sometimes we absolutely do. The sample may be too narrow. We may not have explored the question deeply enough. The evidence may genuinely be incomplete. But sometimes we already know quite a bit. The disagreement is really about what the evidence means or what acting on it would require.

A finding might challenge an existing strategy. The solution might require a capability the organization does not have. Acting might introduce more risk than people are comfortable taking. In those situations, additional data cannot necessarily resolve the underlying tension. The more useful conversation may be: What did we actually learn? What are we assuming that learning means? Why are we favoring one interpretation over another? And what would we conclude if the answer did not have to fit neatly into what we already know how to do?

Those are not simply research questions. They are organizational questions. A simple discipline can help make this visible. I try to separate three things:

• What we heard or observed. The words, behaviors, patterns, measurements, and other evidence that came directly from the research.

• What we believe it means. The explanation we are placing around that evidence.

• What must remain visible. The underlying human need or reality that should continue guiding the work, even if the eventual solution changes.

That last distinction is especially important. Products can evolve. Strategies can shift. Technical approaches can change. But the work can keep progressing while becoming less relevant if the original human need quietly disappears along the way. The value of research therefore does not come only from collecting good evidence. It also depends on our ability to remain clear about what the evidence actually tells us, where interpretation begins, and what must remain visible as decisions are made.

A team can have excellent research and still make a weak decision if context has been lost. It can have a sophisticated dashboard and still misunderstand the people represented by the data. The gap between evidence and action is sometimes a communication problem. But sometimes it is an interpretation problem. The evidence did not change. What people believed it meant did. And recognizing that difference can change the conversation.