The conversion gap is often a confidence gap.

A drop-off is a location, not an explanation.

Analytics can point to the moment someone leaves a journey. That is useful evidence, but it cannot, by itself, explain the decision. Two people can abandon the same screen for very different reasons: one cannot complete the task, while another no longer wants to.

Effort and uncertainty need different remedies.

If someone cannot find the next step, a clearer interface may help. If they cannot tell what they are committing to, removing a field will leave the real concern untouched. This distinction matters because teams can spend weeks optimising the mechanics of a journey while the unanswered question stays in place.

Listen for the question behind the hesitation.

Consider a person opening an account. They may wonder whether they can withdraw easily, how their information will be used, or whether the service meets their needs. In ecommerce, the question may concern returns. In a subscription, it may concern cancellation. The interface should answer the relevant question where it arises, rather than assuming a generic trust badge will do the work.

Research the moment, not just the page.

A useful conversation reconstructs what someone expected before arriving, what they noticed, and when their confidence changed. Ask them to describe a recent decision in their own words. Avoid explaining the experience or defending a design while they are telling the story. Where consent and access allow, combine this with usability observation and funnel data.

Turn the explanation into an experiment.

Write a hypothesis that connects the proposed change to a specific barrier. For example: making the cancellation process clear before trial signup may help people who fear an unwanted commitment. Choose an outcome and a guardrail before launching the test. A higher signup rate is less useful if it produces confusion, complaints, or immediate cancellations.

A better question for the next review.

Alongside “How can we make this easier?”, ask “What does someone still need to understand or believe at this point?” The first question improves the task. The second can improve the decision.