What is eNPS?
Employee Net Promoter Score, or eNPS, measures employees’ stated likelihood of recommending their organization as a place to work. It uses a 0–10 question and subtracts the percentage of detractors from the percentage of promoters.
It is a concise measure of workplace advocacy, not a complete measure of engagement, wellbeing or the reasons people stay. A comment can help explain a response, but neither a score nor a comment establishes an employee’s future behavior.
Use eNPS as one part of an employee-listening program. Plan who will review the feedback, how employees’ information will be handled and what happens after the findings are shared. A survey has limited value if the organization has no capacity to respond.
How is eNPS calculated?
Classify valid ratings as promoters (9–10), passives (7–8) and detractors (0–6). Then calculate: eNPS = percentage of promoters − percentage of detractors. Passives remain in the valid-response denominator. The result ranges from −100 to +100.
For a fictional survey with 100 valid ratings, 45 promoters, 35 passives and 20 detractors produce eNPS of 25. Report it as a score of 25, not “25% engaged.” The measure does not mean that the other employees are disengaged.
Bain discusses employee NPS as part of its employee feedback approach. See Bain’s employee NPS overview. Keep the calculation separate from broader claims about what the metric predicts.
A practical eNPS question set
Scroll horizontally to see all columns →
| Question | Format | Use |
|---|---|---|
| How likely are you to recommend this organization as a place to work? | 0–10 with clear likelihood endpoints. | Workplace recommendation rating. |
| What is the main reason for your rating? | Optional open text. | The employee’s explanation. |
| What is one change that would improve your experience at work? | Optional open text, if it adds useful information. | A practical suggestion. |
Test the wording and explanation before launch. Do not assume every low rating describes a manager problem or every high rating means the employee has no concerns. Avoid asking for unnecessary personal details in comments.
Choose additional questions around a defined need. A broader engagement survey may examine several aspects of work that a recommendation question does not measure. Combining a short pulse with other listening channels can be useful; a frequent eNPS is not automatically a substitute for deeper inquiry.
Explain anonymity and confidentiality accurately
Decide what information is collected, whether responses can be linked to employees and who can access raw responses. Then explain the actual arrangement. “Anonymous” and “confidential” are not interchangeable promises.
If an administrator can connect a response to a person, describe the access and reporting controls rather than promising that identification is impossible. If the design is anonymous, do not create a hidden employee history behind it. Evaluate the full collection setup, including links, imported attributes and exports.
Even without a name, a comment can reveal someone through a specific event, role or phrase. Group-level reporting alone does not remove that possibility. Review what will be shared and whether a paraphrased theme is more appropriate than a verbatim quote.
Explain the process for concerns that require another channel. An employee-listening survey should not be the only route for an urgent issue or an individual request for help. Keep the invitation clear about the survey’s purpose and who will see the feedback.
When should you report eNPS by team?
Team or location views can identify differences hidden in an organization-wide score. They can also expose people in small groups or encourage unsupported comparisons. Choose reporting groups and disclosure rules before reviewing the results.
Do not rely on a single universal minimum count. Consider the group size, response count, identifying attributes and the content being shown. Several filters combined—such as team, tenure and role—may reveal more than each filter alone. Check whether adjacent reports allow a hidden value to be inferred by subtraction.
When a group cannot be reported appropriately, combine it with a suitable larger group or omit the breakdown. Do not create an unsafe view simply because a manager asks for it. Keep access rules aligned with the commitments made to employees.
Also consider whether comparisons are meaningful. Teams may differ in role, location, work conditions and response coverage. A ranking of small groups can imply a level of certainty the data does not support.
Worked example: show score and participation together
Suppose 200 employees are eligible and 100 provide valid ratings. There are 45 promoters, 35 passives and 20 detractors, giving eNPS of 25. Participation is 100/200, or 50%, using that eligibility definition.
Sixty respondents leave usable comments. Twelve mention workload. That is 20% of usable comments, not 20% of the workforce. People may mention several issues, and those who did not comment may have different experiences.
In the next wave, eNPS rises to 30 but only 70 employees respond. Report both facts. The result may reflect change in experience, who responded or other influences. Do not declare that workplace conditions improved solely because the headline score increased.
If the survey is anonymous, analyze the repeated group results within that design. Do not claim that particular employees moved from detractor to promoter. If a confidential longitudinal design is deliberately used, explain its purpose and controls before collection.
Read employee comments without overstating them
Use clear topic definitions, preserve mixed views and check a sample of automated classifications. A comment can express pride in the organization and frustration with workload at the same time. Forcing it into one positive or negative category can lose the useful detail.
Separate what people reported from the team’s interpretation. “Several comments described unclear priorities” is a finding about the feedback. “Unclear priorities caused the eNPS decline” is a stronger claim that needs additional evidence.
Review frequency alongside seriousness and context. A rare concern may need attention even if it does not appear in the top themes. Use appropriately authorized reviewers and avoid circulating raw comments more widely than necessary.
A summary with source references still needs disclosure review. The ability to inspect a passage internally does not mean every report recipient should see it. Verify that permissions also apply to exports and AI-assisted answers.
How often should you run eNPS?
Choose a cadence around the decisions and the time needed to act. A quarterly pulse may fit one organization; another may need a different schedule or a targeted follow-up after a specific change. More frequent collection is not useful if findings simply accumulate without review.
Keep wording, eligibility and collection conditions consistent where comparison matters. Document changes and major events that may influence responses. Avoid repeatedly asking the same question before employees have seen what happened after the previous survey.
Use other channels for dialogue and clarification. CIPD emphasizes listening and acting on employee feedback as part of meaningful employee voice. See its employee voice guidance. A score should support that work, not replace it.
Return a useful response to employees
Share a summary that employees can understand: what the team heard, what it will act on, what needs further investigation and what cannot currently be changed. Explain the next review point. Avoid presenting only the positive themes.
Give each agreed action an owner and a way to check progress. If employees describe unclear priorities, the action might involve a clearer planning process and a later check on whether it is useful. The next eNPS score can provide context, but it should not be the only test.
Do not target an individual because of a low rating or try to identify anonymous respondents. Keep the purpose focused on improving the workplace and understanding experiences. A survey label such as “detractor” should not become an employment judgment.
Build a self-managed employee-listening workflow
Define a small common set of fields and reporting rules across the organization. Allow relevant local questions where needed, without combining different meanings under one label. Record question versions, group definitions and changes in access or disclosure rules.
Sopact’s collection, analysis and governance approach is relevant when the team needs to maintain recurring evidence with clear ownership. Verify the proposed setup against your actual confidentiality commitments and reporting requirements; do not assume grouping or AI analysis makes sensitive feedback safe automatically.
Use a pilot with fictional or appropriately authorized data. Include a small group, identifying comment, changed team attribute, missing response and restricted reviewer. Check the report, export and assistant views before collecting real employee feedback.
For buying requirements, continue to employee survey software. The employee-listening course linked below provides the practical path for planning collection, review and follow-through.
Watch: reviewing incoming feedback
This general feedback video introduces reviewing responses with their context. For employee feedback, apply the confidentiality and disclosure rules above rather than assuming every response should be attached to an identifiable employee.
Frequently asked questions
Does eNPS measure employee engagement?
It measures stated willingness to recommend the organization as a workplace. It can contribute to an employee-listening program but does not cover every dimension of engagement.
Are passives included in the calculation?
Yes. Passives remain in the valid-response denominator. eNPS subtracts the percentage of detractors from the percentage of promoters.
Can eNPS predict who will resign?
A score alone does not establish an individual’s future behavior. Use feedback to understand experiences and support appropriate organizational review, not to label individual employees as likely departures.
Is a group-level comment report automatically anonymous?
No. Details in a comment or a combination of filters can identify someone. Review the collection, access and disclosure design before making an anonymity promise.
Should every manager receive a team score?
Only when the reporting design supports a useful and appropriate view. Small groups, response coverage and identifying details may require a broader grouping or omission.
What should we do after the survey?
Review the evidence, choose practical actions, assign owners and tell employees what will happen next. Check the issue itself rather than treating a later score increase as proof of resolution.
Choose your next step
King Center
Related practice: reviewing participant feedback across programs. This story does not demonstrate employee-survey confidentiality controls.
Read the customer story →