eNPS is the employee net promoter score: one question, one number, and more argument per data point than any other metric in HR. Some organisations run their entire engagement programme off it. Others dismiss it as a vanity figure. Both positions are defensible, which tells you the number is being asked to do more work than it can carry.
This guide covers what eNPS measures, how to calculate it, what a good score actually looks like, and — the part that decides whether it is useful — what to do with it once you have one.
Table of Contents
What eNPS is
eNPS adapts the Net Promoter Score, introduced by Frederick Reichheld in a 2003 Harvard Business Review article, from customers to employees. The customer version asks whether you would recommend the company to a friend. The employee version asks whether you would recommend it as a place to work.
The standard wording is: On a scale of 0 to 10, how likely are you to recommend this organisation as a place to work?
Responses split into three groups:
- Promoters (9–10) — actively positive, likely to advocate and to stay
- Passives (7–8) — content enough, not committed, excluded from the calculation
- Detractors (0–6) — would warn someone off working here
The 0–6 detractor band surprises people. A 6 out of 10 sounds like a pass; in this scoring it counts against you exactly as heavily as a 0. That is deliberate — the scale is calibrated to advocacy, and lukewarm is not advocacy.
How to calculate eNPS
The formula is:
eNPS = % promoters − % detractors
Worked example. You survey 200 people and 160 respond:
- 72 promoters = 45%
- 56 passives = 35%
- 32 detractors = 20%
45 − 20 = an eNPS of +25.
Percentages are calculated from people who answered, not people you asked. The range runs from −100 to +100, and the result is a score, not a percentage — writing “our eNPS is 25%” is a small error that undermines the rest of the presentation.
Note what the maths does to passives. They dilute both percentages without appearing in the subtraction, so an organisation where almost everyone sits at 7 or 8 posts a score near zero. That looks like a crisis and is usually something much duller: a workforce that is fine and not especially moved.
What is a good eNPS score?
The rough consensus across published benchmarks:
| Score | Generally read as |
|---|---|
| Below 0 | More detractors than promoters — needs attention |
| 0 to 10 | Neutral; common in large or heavily process-driven organisations |
| 10 to 30 | Healthy |
| 30 to 50 | Strong |
| 50+ | Exceptional |
Now the caveat that matters more than the table. Published benchmarks disagree with each other badly — global averages cited in reputable sources range from about 12 to about 27 depending on who is counting, which sectors are in the sample and which quarter it was taken.
Organisation size skews it further. Smaller employers routinely post averages around 30 while the largest sit closer to 9, so a score of 20 can be an underperformance for a 60-person firm and a genuine achievement for one with 8,000 staff.
The practical conclusion: your own trend is a far better benchmark than anyone else’s average. A score moving from +8 to +19 over three quarters tells you something real. The same +19 compared against a global figure of unclear provenance tells you almost nothing.
What eNPS is good at
- It is short. One question, high completion, minimal fatigue. You can ask it often without the response rate collapsing.
- It travels. Boards understand a single directional number. Getting engagement onto the board agenda is easier with eNPS than with a 24-dimension framework.
- It is sensitive. Because passives are excluded, small shifts in genuine advocacy move the score visibly. That makes it good at catching change early.
- It segments well. The interesting number is never the company one. It is the gap between two departments, or between people with under a year’s service and everyone else.
What eNPS cannot tell you
It cannot tell you why. That is not a flaw in the metric, it is the definition of a single-item measure — but it is where most eNPS programmes quietly fail. A score drops eleven points, everyone agrees this is concerning, and nobody can say what happened, so the meeting resolves to watch it next quarter.
Two scores of +20 can also describe completely different organisations. One with 60% promoters and 40% detractors is polarised and volatile. One with 25% promoters, 70% passives and 5% detractors is stable and flat. Same score, opposite problems, opposite responses. Always publish the split alongside the score.
It is also blunt at small numbers. Below roughly 30 respondents, one person shifting from 7 to 6 moves the score several points, and team-level scores in a small business are mostly noise. Reporting it weekly to a team of eight will teach that team to distrust data generally.
The question that makes eNPS worth asking
The score is the thermometer. The follow-up is the diagnosis.
Always pair the rating with an open follow-up — “what is the main reason for your score?” — and read the comments by band. Detractor comments tell you what to fix. Promoter comments tell you what to protect, which organisations forget to do until they have accidentally dismantled it. Passive comments are the most underrated of the three, because passives are the people whose minds are still changeable.
This is also where anonymity stops being a nicety. People will not write the honest version of the reason if they think it is attributable, and a comment field nobody trusts produces a dataset of “nothing to add” — a dynamic we have covered in whether engagement surveys are really anonymous.
How often to ask
Annually is too rare to act on and quarterly is the common default, but the honest answer is that cadence matters less than continuity. An eNPS asked continuously to a rotating sample gives you a rolling line rather than four dots a year, and a line is what lets you connect a movement to the restructure, the policy change or the manager departure that caused it.
Four dots a year cannot do that, which is the broader case we have made about the annual engagement survey.
It is how we built eNPS in wotter: asked continuously rather than in a quarterly burst, reported with the promoter and detractor split intact, and sitting alongside the 24 Core24 pillars — so when the score moves, you can see which part of the experience moved with it.
Five ways eNPS programmes go wrong
- Making it a manager target. Tie bonuses to eNPS and you will get a better score and worse information. Goodhart’s law applies to engagement metrics with unusual speed.
- Changing the wording. Rephrase the question and the trend line breaks. Keep it identical even when it feels stale.
- Reporting the score without the split. See above — the split carries the meaning.
- Chasing the benchmark. Optimising against a global average of contested provenance is not a strategy.
- Asking without responding. The fastest way to depress next quarter’s eNPS is to collect this quarter’s and do nothing visible with it.
Used as one signal inside a broader listening approach, eNPS earns its place. Used as the whole programme, it becomes a number your leadership team watches and nobody can move.
Frequently asked questions
Quick answers on scoring, benchmarks and the eNPS question itself. Wotter is an employee engagement platform — we track eNPS continuously alongside 24 engagement pillars, so a change in the score arrives with a reason attached. Anything else, drop us a line at hello@wotter.group.
