eNPS: what it does and doesn't tell you

eNPS (employee Net Promoter Score) asks how likely people are to recommend your organisation as a place to work, on a scale from 0 to 10, and subtracts the share of detractors (0 to 6) from the share of promoters (9 to 10). It tells you, quickly and cheaply, whether sentiment is moving and where. It doesn’t tell you why, and the number on its own is easy to misread.

How is eNPS calculated?

Ask one question on a 0 to 10 scale. Bain, which created the method, says most adopters use a version of “How likely would you recommend this company as a place to work?”

Then sort the answers into three groups, using the same bands as the customer score on Bain’s Net Promoter System site:

• Promoters answer 9 or 10.
• Passives answer 7 or 8.
• Detractors answer 0 to 6.

eNPS is the percentage of promoters minus the percentage of detractors. Passives count towards the total but not the score. Say 200 people answer: 70 promoters (35%), 80 passives (40%) and 50 detractors (25%). Your eNPS is 35 minus 25, which is +10. The score can run from −100, where everyone is a detractor, to +100, where everyone is a promoter.

Where does eNPS come from?

The Net Promoter Score was built for customers. Fred Reichheld of Bain & Company introduced it in the Harvard Business Review article “The One Number You Need to Grow” in December 2003. Bain’s own history of Net Promoter credits a research team at Bain led by Reichheld.

Companies later turned the question inward. Bain’s employee NPS page names Apple among the adopters and says the likelihood-to-recommend question is the best predictor of retention it found. It also suggests a second question, on whether people find their job inspiring, which it links to energy and enthusiasm. And it warns that employee scores can be substantially lower than customer scores, because employees tend to judge their own company more harshly.

What can eNPS tell you?

• Whether sentiment is moving. One question can be asked often without wearing people out. Bain says companies most commonly ask monthly.
• Where to look. Compared across your own departments, sites or shifts, it shows which groups sit well below the rest.
• Whether something you changed made a difference, if you measure the same group before and after.

Used this way, eNPS is a good smoke alarm. It is cheap, simple to explain to a leadership team and easy to track on one chart.

What doesn’t eNPS tell you?

Why people gave the score

A driver at a depot who answers 5 might be unhappy with pay, the roster, a manager, a broken heater in the break room or the commute. The number can’t distinguish these. The reasons are what you would act on, and the score has none of them.

How the answers are spread

Very different teams can share a score. A team of 20 promoters, 60 passives and 20 detractors scores 0. So does a team of 50 promoters and 50 detractors. The first is lukewarm. The second is split in two, perhaps between day and night shifts or between two sites, and needs a completely different conversation.

What is happening to the passives

Passives vanish from the calculation, yet they are often the biggest group. A slide from 8 to 7 doesn’t change the score. A slide from 7 to 6 does, and by then people may already be looking.

Anything reliable about small teams

In a team of ten, one person moving from 9 to 6 shifts the score by 20 points. Read small-team scores as a prompt to ask, and avoid ranking managers on them.

What is a good eNPS score? Why benchmarks are risky

Published eNPS “benchmarks” vary widely from one vendor report to the next, and comparing your score with them is risky. The question wording differs. So does who was asked, when, how anonymous it felt, which industries and countries are in the sample, and whether the survey was voluntary. A 9 means different things to different people.

The comparison that holds up is with yourself: the same question, asked the same way, to the same groups, over time. And between your own teams, read side by side.

How do you find out why people gave the score?

Customer NPS programmes often call detractors back. Bain says that with employees this isn’t possible, because employee feedback must be kept confidential to keep it honest. So you follow up with groups, not individuals.

1. Add one open question straight after the score: “What is the main reason for your answer?” You will get short replies, but they help you sort the first themes.
2. Find the groups that sit well below the rest or that moved most since last time.
3. Ask the people in those groups, in a format where they can explain. One-on-one conversations work better than focus groups when the reason might involve a manager or a colleague.
4. Report back what you heard and what you will change, then measure again.

Spradley does step three at scale. Each person gets a one-on-one AI interview by text or voice. It can ask the recommend question in plain words, “Would you recommend working here to a friend? What would you tell them?”, and the interviewer follows up on each answer. The report groups the reasons into themes with quotes, by department, and never shows numbers or ratings for fewer than five people. See the employee engagement page for how it works.

People will only give the real reason if they feel safe doing so. If you suspect they don’t, our post on how to measure psychological safety is a good next read. If your eNPS has just fallen, start with why engagement scores dropped, and if you are weighing eNPS against other methods, see alternatives to the annual engagement survey.

Frequently asked questions

Key terms

eNPS

Employee Net Promoter Score. People rate how likely they are to recommend the organisation as a place to work from 0 to 10, and the share of detractors (0 to 6) is subtracted from the share of promoters (9 to 10).

Net Promoter Score

A customer loyalty measure introduced by Fred Reichheld of Bain & Company in 2003, based on how likely people are to recommend a company. eNPS applies it to employees.

promoters, passives and detractors

The three groups in an NPS or eNPS calculation: promoters answer 9 or 10, passives 7 or 8, and detractors 0 to 6. Passives count in the total but not in the score.

benchmark

An external comparison figure, such as an average score across an industry. Differences in wording, sample and timing can make benchmarks misleading.

minimum group size

The smallest number of responses a group needs before its results are shown. Five is a common default; smaller groups are hidden or combined with a larger one.