Last CallLast Call Pool
Research & Discovery

Insight driven product - how to ask users and don't suggest answers?

Power Talk
🇵🇱 Polish

About

Even the most experienced researchers make mistakes in qualitative research like in-depth interviews or usability testing. It can be nodding when a user says what you want to hear, not allowing silence to let them think, or assuming they may be emotional about a product because you feel they might be facing a crisis.

How can psychology help here? What is understood by empathy? How to focus on what the other person is trying to tell you and not plan the next question in your head? How to train "soft" skills, and where to draw inspiration to enhance them?

The presentation is a collection of practical tips (with examples of ups and downs) on qualitative research and mitigating mistakes. I will cover the most common situations and questions a researcher gets asked. The speech is a sort of answer to the question: how to interview users and make as few suggestions as possible?

As it is no longer only UX Researchers who interview users, the talk is addressed to Product Owners, Project Managers, Product Designers, and UX Designers. I will start with the basics of interviewing and testing and move on to more advanced cases and a toolbox for strengthening competencies in this area.

‍

Talk by

Photo of Joanna Gierasimiuk, Senior UX Designer & Researcher at Sparing Digital
Joanna Gierasimiuk
Senior UX Designer & Researcher at Sparing Digital

Watch the full talk

Watch this WaysConf session, then continue with related talks or explore the current programme.

Explore the 2026 agendaMore WaysConf recordings
From the recording

Talk in brief

A practical guide to asking better questions in product research. The talk explains why direct questions about future behavior produce declarations rather than evidence, why participants should be allowed to encounter confusion without immediate rescue, and why roadmap questions cannot be answered by people who lack delivery context. Good interviewing requires neutral prompts, observation of past behavior, triangulation across sources, and a setting that does not pressure participants to please the researcher. The goal is useful truth, not a smooth conversation.

Key takeaways

  1. 01

    Ask about evidence instead of intentions

    Future-use questions invite optimistic declarations, while recent behavior and concrete examples reveal stronger product evidence.

    Watch from 6:31
  2. 02

    Do not outsource roadmap promises to users

    Participants can explain needs and consequences, but they do not know the team’s technical plan or release constraints.

    Watch from 9:18
  3. 03

    Let confusion remain observable

    Explaining an interface too quickly removes the moment researchers need to understand where the product fails.

    Watch from 12:19
  4. 04

    Triangulate findings across several sources

    Interviews, behavior, analytics, and other evidence can challenge one another and reduce confidence in a preferred story.

    Watch from 15:32
  5. 05

    Reduce pressure to please the interviewer

    Hospitality and leading reactions can encourage polite answers, so researchers should create permission for criticism and compensate participation fairly.

    Watch from 18:34
Read edited transcript highlights

These concise notes were edited from automatic captions and checked against the talk structure. They are not a verbatim transcript.

Past behavior is stronger than stated intent

Asking whether someone would use a proposed solution requests a prediction and signals the answer the team hopes to hear. Instead, investigate the most recent occasion when the underlying need occurred. Ask what happened, what the person tried, what it cost, and what they did next. Concrete history does not guarantee the future, but it provides observable evidence.

Watch from 6:31

Participants explain needs, not release plans

Customers can describe urgency, workarounds, and the consequences of delay. They cannot determine when a feature should leave beta because they do not know dependencies, priorities, or quality thresholds. Researchers should capture the need accurately and avoid implying a delivery promise that belongs to product and engineering decision-makers.

Watch from 9:18

Do not remove the evidence by helping too soon

When a participant hesitates, a researcher may instinctively explain the product and restore a comfortable conversation. That intervention erases the point of observation. Allow enough silence for the person to interpret the interface, then ask what they expected and what they are looking for. Support can follow after the behavior has been understood.

Watch from 12:19

Conflicting sources are valuable research output

Interviews may suggest one story while analytics, support records, or observation suggest another. This is not a reason to select the most convenient source. Trace the difference through sample, context, and behavior. Triangulation limits confirmation bias because the conclusion must account for multiple forms of evidence rather than one compelling quotation.

Watch from 15:32

Social comfort can distort honest feedback

Participants often try to be helpful, especially when the interviewer reacts enthusiastically or creates a sense that criticism would be impolite. Use neutral acknowledgment, state clearly that the product rather than the person is being tested, and compensate the participant appropriately. A pleasant atmosphere is welcome, but credible evidence remains the purpose.

Watch from 18:34
Explore WaysConf 2026

View the agenda

Get tickets