What this section is for
Ask developers what helps or hinders their work: tools, processes, workload and collaboration. Create a template with ratings, agreement scales, NPS, choices or free text. Run it for selected dates. Results become available once enough responses have been collected.
When to use
- For regular team feedback, such as once a quarter.
- Before and after a substantial change in tools or process.
- To learn about difficulties that repository data does not show.
Quick start
- 1Build a template, group questions into categories, and keep it short enough that people finish it.
- 2Start a cycle with a start and end date, decide on anonymity, and set the minimum number of responses required before results appear.
- 3Share the survey link. Participation percentage is available only for a known, non-anonymous audience.
- 4Once enough responses arrive, review categories and individual questions. Identify which comments need follow-up.
Key settings
| Setting | What it does | How to use it | Common mistake |
|---|---|---|---|
| Anonymous responses | Controls whether responses can be traced back to a person. | Explain anonymity and its limits before sharing the survey: detailed free-text answers can sometimes reveal the author. | Promising complete anonymity while asking for identifying details. |
| Minimum responses for results | Withholds results until enough responses have arrived. | Choose a threshold that prevents inferring an individual response. Account for team size. | Setting the threshold to one and treating that as adequate anonymity protection. |
| Cycle dates | The window during which the survey accepts responses. | One to two weeks, and keep the same interval between cycles so scores are comparable. | Running a cycle during a release crunch and then comparing the result to a quiet quarter. |
Usage example
Scenario: The team wants to understand difficulties at work even though release dates have not slipped.
- 1Run an anonymous survey about tools, process, workload and clarity of tasks.
- 2After the response threshold is met, compare categories and questions. The overall average may hide a particular problem.
- 3Choose a concrete improvement and explain it to the team. Ask the same questions next time to compare answers.
What this shows: You have feedback on specific working conditions and have chosen what to change first.
FAQ
Why can I not see the results?
The cycle has not reached its minimum response count. The threshold exists to protect anonymity and is not bypassed.
Participation is low. What now?
Check that the link works, the survey is still open and the questions are clear and not too time-consuming. Explain who can see the answers and how results will be used. Share what changed after the previous survey.
Which question type should I use?
Use agreement scales for statements, ratings for satisfaction and NPS for willingness to recommend. Free text adds explanations. Keep questions whose answers you can use.
Tips
- The overall score averages valid 1โ5 ratings within each response, then averages across respondents. NPS is separate: the share of 9โ10 ratings minus the share of 0โ6 ratings.
- Each question must meet its own minimum response count. Partial responses are excluded. Participation percentage requires a known non-anonymous audience.
- Category counts are question answers, not unique people. No automatic time trend is calculated; compare cycles with the question set in mind.
- Keep question wording consistent between cycles and tell the team what changed after the survey.