Using Empathy to Cut Through Noise in User Research

Whether in the internet or traditional industries, companies strive to listen to user or consumer voices, yet the process introduces various noises. This article examines three noise sources—user participation motivation, expression willingness, and expression authenticity—and shows how the tool of "empathy" (i.e., perspective‑taking) can be used to reduce them.

User Participation Motivation

Ask yourself: why does a user join a study? If the motive may negatively affect attitudes (e.g., seeking freebies or gathering competitor intel), consider discarding the sample during the user‑screening stage for qualitative research, or set filter questions in surveys for quantitative studies.

Example: A student, Xiao Wang, signed up for a face‑to‑face interview to boost his résumé. After evaluating his motive, the team decided not to invite him.

User Expression Willingness

Consider why a user is willing to speak: personality, incentives, interest, or desire for product improvement. Highly expressive users can dominate data collection, while silent users may be filtered out. For qualitative research, remove overly expressive users during screening; for both qualitative and quantitative research, control their "voice volume".

Examples include:

Introverted users giving brief answers may be unsuitable for deep interviews.

Highly talkative users in usability tests may need gentle interruption to keep focus.

Active product users may dominate questionnaire results; sample weighting or multi‑channel outreach can balance representation.

User Expression Authenticity

Even cooperative users may provide inaccurate data. Common issues are:

Misunderstanding the question (e.g., not grasping the intended meaning).

Lack of self‑awareness about what they can say.

External influences, such as wanting to appear competent or friendly.

To address these, use clear questionnaire wording, bold or colored highlights, and layered questioning. Apply the "5 Why" analysis to dig deeper into user reasoning.

Dialogue Example:

Researcher: What do you think of the product’s description?

User: It’s too cluttered.

Researcher: Which parts feel redundant and why?

User: The “oil‑free” claim isn’t relevant to me.

Sample Weighting and Multi‑Channel Reach

When sample sizes for sub‑groups are sufficient, weight them according to actual user distribution. For an online learning app with 2,000 respondents (30% new users, 70% old users), adjust weights to reflect a 40/60 split, improving representativeness.

If the study also needs potential users, extend recruitment beyond the current user base, e.g., via sister products or sample‑service providers.

Mitigating External Influences

During research, users may feel observed and alter responses. Reduce this by limiting the number of observers, using one‑way glass rooms, or providing warm‑up sessions that reassure participants there are no right or wrong answers.

Control wording to avoid leading language and focus on past behaviors rather than hypothetical future actions.

Conclusion

Key takeaways for noise reduction: evaluate user motivation during screening; adjust voice volume based on willingness; enhance authenticity through clear questions, follow‑up probing, and minimizing external bias. Empathy is a skill that, together with team and product empathy, helps uncover genuine user needs.