What needs to be finished first
A precise research question exists that can meaningfully be answered with structured responses from many people.
Target group and desired sample are defined, including exclusion criteria and recruiting channel.
What needs to be ready before start
Survey tool such as Typeform, Google Forms, Qualtrics or LimeSurvey; questionnaire draft; sample plan; evaluation sheet; consent and privacy text.
Research owner for questionnaire and evaluation; domain reviewer; optional data analyst; participants from clearly defined segment.
Research question, target group, desired n count, decision supported by results, and hypotheses or answer options.
3-14 days
Limit questionnaire to maximum 8-12 minutes. Use required questions sparingly. Pilot with 3-5 people before sending.
The one question this method answers
Which patterns, frequencies or priorities appear in a sufficiently large target group?
Marker: Phase
| Step | Duration | Action | Hint |
|---|---|---|---|
1Operationalize research question | 30-60 min | Define decision, hypotheses and measurable variables. Include only questions that directly contribute to evaluation. | If nobody can explain which decision a question influences, it does not belong in survey. |
2Build questionnaire | 60-120 min | Write clear questions with appropriate answer formats, keep scales consistent and put demographic questions at end. | No double-barreled questions such as "easy and fast". Each question measures exactly one thing. |
3Run pilot | 30-60 min | Test questionnaire with 3-5 people, correct unclear wording and drop-off points. | A pilot saves more time than it costs. Scales and multiple choice are especially often misunderstood. |
4Collect data | 2-10 days | Send survey to target group, check response, send reminders and mark obvious quality problems. | Fast response from a single channel can be biased. Document source of responses. |
5Evaluate and decide | 2-4 h | Clean responses, analyze frequencies and segments, formulate core statements with uncertainty and limits. | Survey results show patterns, not automatically causes. Why questions often need follow-up interviews. |
What comes out at the end
Survey report with research question, sample, field time, questionnaire, cleaned n count, charts, segment analysis, key insights and decision recommendation.
Save questionnaire, field time and raw data as snapshot. On repetition, assign version number and mark question changes as breaking changes.
Compact working template for Surveys with research question, sample, questionnaire, and readout.
Collect structured responses at scale and turn them into measurable insight.
When and for what do we use this method?
Which data, observations, decisions, or materials are available?
Owner, date, and success signal.
Concrete filled scenario, fictional example
Concrete filled scenario, fictional example
Research question: Which steps in client onboarding create most drop-offs? Sample: 142 clients, field time 2026-05-06 to 2026-05-12, n=118 after cleaning.
Top results:
Segment: New founders under 12 months name upload problems much more often at 61%.
Decision: Prioritize upload checklist and progress indicator. Three interviews with founder segment follow to clarify causes.
Recognize symptoms and steer against them
Drop-off rate rises, free-text answers become short or empty.
Cut questionnaire back to decision question and remove nice-to-have questions.
Answer options visibly steer toward preferred solution.
Write neutrally, balance options and let pilot participants think aloud.
Small samples presented with decimals.
Show n count, report rough magnitudes and name uncertainty.
Team interprets motives although only behavior or preference was asked.
Mark causes as hypotheses and deepen with interviews or usability testing.
Only very satisfied or very frustrated users respond.
Document recruiting channel and non-responses, do not sell results as whole market.
Done signals checkable in under a minute
Go to the profile for purpose, similar methods, and sources or continue to the next method in the catalog.