eNPS describes employees’ stated willingness to recommend their organisation as a place to work. It is a signal to investigate, not a complete measure of satisfaction or engagement. Culture Amp explains why a single question cannot replace a multi-item measure.
Question and scale
How likely are you to recommend this company as a place to work?
Use 0–10, anchored from “not at all likely” to “extremely likely”. Keep the wording, scale and invitation method consistent across waves. For international teams, review translations for equivalent meaning before launch.
| Response | Group | Role in the calculation |
|---|---|---|
| 9–10 | Promoters | Numerator and denominator |
| 7–8 | Passives | Denominator |
| 0–6 | Detractors | Subtracted in numerator; included in denominator |
These labels describe responses, not employees’ personalities or performance.
Formula and example
eNPS = 100 × (number of promoters − number of detractors) / number of valid eNPS responses. The range is −100 to +100 points. It is not the average rating.
Illustrative example: 50 promoters, 30 passives and 20 detractors produce 100 × (50 − 20) / 100 = +30. Excluding passives would incorrectly produce about +42.9. A missing answer is not a zero. People who skip eNPS are excluded from its denominator.
Adaptations using other scales exist. Disclose their calculation and do not treat them as interchangeable with the conventional 0–10 measure.
Is +30 a good score?
A positive score means more promoters than detractors among respondents. It does not certify a healthy workplace. Look at valid response counts, participation, the distribution and who took part. A benchmark needs a documented source, period, population and comparable questions and scales.
With 20 responses, moving one answer from detractor to promoter changes eNPS by 10 points. This illustrates sensitivity; it is not a significance test. See interpreting small-team results.
How often should you ask?
Match cadence to purpose. A broader survey once or twice a year can be useful; targeted pulses can run more often when a specific question needs follow-up. More data points are not a goal in themselves. See choosing a survey cadence.
Report new-hire surveys as a separate cohort. Do not silently combine their results with a workforce-wide trend.
Turn responses into a conversation
Add an optional question: “What most influenced your answer?”. Comments add context, not proof of causes, and can identify their authors. Define confidentiality rules before launch.
Choose a small number of actions, assign owners and explain the response to employees. A later score shows a change in answers, not proof that your intervention caused it. Use the post-survey action plan.
Prepare a survey in the platform
Choose an NPS 0–10 question, then review participants, the invitation and report settings. Create an account and prepare your survey, or start with our example question set.
eNPS by background group
Department, location and tenure breakdowns are useful only when they serve a purpose and protect respondents. Calculate eNPS from valid answers in each group. Overall eNPS is not the unweighted average of department scores. Missing background information does not exclude a valid answer from the overall result.
Before reporting, check counts, combinations and reconstruction of small groups from totals. See the group calculation example and GDPR and privacy guidance.
New ready-made platform presets require an eNPS score from 0 to 10 before submission; comments remain optional. This is a product setting, not a universal methodological standard. Participation remains voluntary. Requiring a score may increase survey abandonment, which should inform interpretation of response rates. Surveys already launched retain their recorded settings. A missing answer remains missing data, not zero.
Sources and scope
These sources support the principles discussed here. Examples and schedules are starting points to adapt, not standards or evidence that our templates have been validated.
