UX Researcher

The ability to communicate research findings against organizational resistance: a clear presentation of the evidence, a distinction between strong and weak evidence, consultation of the relevant stakeholders before broad communication, patient repetition with different conveyance formats (report, highlight video, workshop). Bonus: the candidate names a case where the initial skepticism was justified and the study had to be methodically sharpened. Anyone who describes a pure victory over the hierarchy shows a black-and-white attitude instead of mature stakeholder competence.

Openness about their own methodical weaknesses: a separation between a method error (wrong sample, a poor task, a too-leading guide) and an execution error (recruiting mishaps, a tech failure). Bonus: the candidate names a concrete adjustment to their own practice (a new pilot step, a different sampling strategy, a different recruiting tool). Anyone who has never reflected on a study methodically has either not run enough studies or reflects superficially.

A partnership posture instead of research-as-veto: a willingness to understand the product view before deciding; a proposal for a leaner study design that balances pace and depth of insight; an acknowledgment when the study was objectively not the most important lever. Candidates who describe PMs or designers as not understanding research show a weakness in cross-functional teamwork that becomes especially expensive at an SMB with a small product team.

Clarifying the decision before the method choice: not agreeing directly, but understanding which decision is pending, who the stakeholders are and which evidence is actually decision-relevant. Expected: 2-3 structured questions, then a proposal for a fitting study design (5 interviews can be right; they can also be too few for a roadmap decision or too many for a pure hypothesis validation). Anyone who simply agrees and runs interviews in 5 days without checking the decision logic shows an executing instead of an advising posture.

Factual clarification instead of escalation: first understand how the team landed on the old persona (an onboarding gap, no update, poor findability), then assess the implication of the shift for the pending decision. Bonus: the candidate names a lightweight update path (a quick sync with the team, a fast validation study, an updated document in the repository). Anyone who reflexively insists on the obsolescence without offering the update path shows a pure guardian posture.

Pragmatic prioritization: not trying to fix everything at once. Expected: (1) an audit of the existing findings and identification of the 10-20 most valuable studies, (2) a choice of a lean repository tool suited to the team size (Dovetail, a Notion database, EnjoyHQ; not straight to the enterprise solution), (3) a convention for atomic research insights and tagging, (4) a joint migration of the most important studies with designers and PMs to secure adoption. Anyone who immediately calls for a full Dovetail enterprise implementation with an elaborate taxonomy shows a lack of pragmatism.

A structured approach: (1) sharpen the activation definition (step A, B or C?), (2) set up a mixed-methods plan: a quantitative analysis of the drop-off points from the analytics data, qualitative depth at the breakpoints with usability tests or diary studies, (3) a hypothesis list of perceived hurdles (friction in signup, an unclear value proposition, missing aha moments), (4) a method choice per hypothesis with an explicit effort-vs-insight trade-off, (5) a delivery plan in 2-week iterations with continuous synthesis instead of a big-bang report at the end. Anyone who jumps into a pure series of usability tests or a 200-person survey without framing the mixed-methods logic shows a method weakness.

Clarification before the proposal (available data, trial volume, the ability to re-contact the drop-offs, the time budget for the study). A coherent plan: (1) a quantitative pre-analysis of the drop-off profiles from analytics, (2) re-contacting a sample of 8-12 drop-offs for deep interviews or short surveys, (3) in parallel 5-8 usability sessions with active trials to observe behavior before abandonment, (4) synthesis into hypothesis clusters with a clear separation between product hurdles and expectation misfit. Bonus: acknowledging the self-selection bias in re-contact (whoever comes back for an interview is not representative of the silent drop-off share). Anyone who proposes a 100-person survey immediately without clarification shows a quantitative reflex posture instead of a method-aware diagnosis.

A structured synthesis method: first recap the research question (what was to be found out?), then sort the observations (what is behavior vs. opinion vs. wish?), then identify patterns across several interviews (what recurs, what is a one-off?), finally derive clear hypotheses or recommendations with evidence strength. Bonus: the candidate explicitly names what the data CANNOT prove, and guards against over-interpretation. Anyone who inflates single statements into insights or reflexively builds a persona from them shows a superficial synthesis competence.

An understanding of saturation logic: 5-8 participants per homogeneous segment for usability tests, 8-15 for exploratory interviews with thematic saturation, more for heterogeneous use cases or multiple segments. Bonus: the candidate names a case where saturation was reached earlier than planned, and they cancelled or repurposed the last planned sessions; or conversely, where they adjusted because a new pattern emerged. Anyone who answers 10 people for everything in a blanket way or cannot explain saturation shows weak method foundations.

A solid tool landscape with a clear rationale: recruiting (Respondent, User Interviews, Prolific, an own panel; depending on the use case), running (Lookback, Maze, dscout, Zoom with a structured guide), analysis (Dovetail, Reduct, a Notion database for lightweight setups), survey (Typeform, Qualtrics, SurveyMonkey). Bonus: the candidate distinguishes between their personal preference and the tool choice that fits the team size and budget. Anyone who answers Dovetail for everything in a blanket way without reflection, or rattles off a tool list without justification, shows weak experience with ResearchOps trade-offs.

Pragmatic conveyance: staggered formats by audience and detail depth (a highlight video or slide deck for the leadership level, a detailed report for the implementing team, atomic insights in the repository for later findability), live workshops for strategic studies, asynchronous updates for ongoing studies. Bonus: the candidate names a continuous insight-sharing practice (a weekly highlight Slack post, a monthly research-review session, mandatory onboarding reading for new joiners). Anyone who delivers 50-page PDF reports as the standard or expects the team to pull findings from the repository themselves has poor prospects at a tech SMB.

Openness: the ability to separate methodology from personal ego. Bonus: the candidate names a case where a valid methodical critique actually led to adjusting a study. Anyone who describes defending their methodology to the critic instead of listening shows a coachability weakness; at an SMB with a small research or design team, that is a hard knock-out signal.

An active enablement posture: pair sessions on interviews, lightweight guide templates for DIY research, regular internal training on interview basics and bias avoidance, review of studies run by non-researchers. Anyone who says research belongs in my hands without anything more concrete shows a gatekeeper attitude that does not scale at an SMB with a small research team. The ability to build a research culture beyond your own team is decisive for the company's research maturity.

Acknowledging the developments: the rise of generative AI in synthesis (automatic transcription, AI-assisted tag suggestions, cluster suggestions), a shift from study-as-project to continuous discovery, the maturing of ResearchOps and repositories, growing pressure on measured study impact instead of pure study production, the democratization of DIY research by PMs and designers. Candidates who only talk about tools or buzzwords show a superficial attitude; anyone who speaks of the tension between pace and depth, of the researcher's role as an enabler rather than just an executor, and of empirical discipline in the AI age is current.