How do you calculate eNPS?

Ask every employee the standard question on a 0–10 scale. Group responses: Promoters scored 9–10, Passives 7–8, Detractors 0–6. Divide each group by total respondents to get a percentage. eNPS = %Promoters − %Detractors. The result is a whole number between −100 (everyone is a Detractor) and +100 (everyone is a Promoter). Passives are excluded from the formula but included in the base count — a large Passive cohort lowers both percentages, which dilutes the score over time.

What is a good eNPS score?

Any positive score means more Promoters than Detractors, which is a reasonable baseline. The published interpretation bands are: 0–10 positive, 10–30 good/healthy, 30–50 strong, 50+ excellent. The global average sits at 12–14 (Perceptyx, 20M+ employees, 2024) or 32 (QuestionPro 2025) depending on which dataset you use — both are vendor-reported and skew toward their client base, so always benchmark within your industry and company-size band.

How often should you run eNPS?

Quarterly is the most common professional cadence — it aligns with planning cycles and gives time to act on results before the next wave. Some organizations run it monthly, but only if they have the capacity to analyze open-ends and share back within two weeks each time. Running eNPS faster than you can act on it trains employees to ignore the survey. Annual is too infrequent — you will miss meaningful trend movement for up to 12 months.

What is the difference between eNPS and an employee engagement survey?

eNPS is one number that measures loyalty and advocacy — whether employees would recommend the organization as a place to work. An engagement survey measures the drivers of discretionary effort across multiple dimensions (clarity of expectations, recognition, growth, management quality, and so on). eNPS tells you the outcome; an engagement survey tells you the causes. Use eNPS as a fast recurring triage signal, and run a comprehensive engagement survey when you need to diagnose root drivers and plan specific interventions.

Can you use eNPS for small teams?

You can run the survey with small teams, but you should suppress results for any subgroup smaller than five respondents. In a team of four, one Detractor score represents 25% of headcount and can be identifiable, which destroys the honest-response environment the survey depends on. Set and communicate your minimum reporting threshold before launch. For very small organizations (fewer than 20 people), consider whether a confidential facilitated conversation delivers better data than a numerical score.

Employee Surveys

What Is eNPS and How Do You Calculate It?

eNPS (Employee Net Promoter Score) is calculated by asking every employee one question on a 0–10 scale, grouping responses into Promoters (9–10), Passives (7–8), and Detractors (0–6), then subtracting the percentage of Detractors from the percentage of Promoters — passives are excluded from the formula but counted in the base. Scores range from −100 to +100; the global average sits around 12–14 (Perceptyx, 20M+ employees, 2024) or 32 (QuestionPro 2025), depending on dataset. A score above 0 is positive, 10–30 is considered good, 30–50 is strong, and 50+ is excellent.

5 QuestionseNPS / LoyaltyQuarterly0–10

Score It

Calculate Your eNPS

eNPS calculator

eNPS (Employee Net Promoter Score) measures employee loyalty on a single 0–10 question. Respondents are grouped into Promoters (9–10), Passives (7–8), and Detractors (0–6). Passives are excluded from the formula but included in the denominator, so a high Passive count still dilutes your score over time. The result is a whole number between −100 and +100. Formula: % Promoters − % Detractors.

Promoters (9–10)Loyal advocatesPassives (7–8)Satisfied, unenthusiasticDetractors (0–6)At-risk / unhappy

Total responses: 0

Your eNPS

—

Enter response counts to score

The Question Bank

Copy-Ready Questions, Grouped by Theme

Every group uses the scale that fits it. Copy one question, a whole theme, or the full set straight into your survey tool.

Filter by theme5 questions

The eNPS Question

0–10 scale0 (not at all likely) → 10 (extremely likely)

The single rating item that anchors the eNPS calculation. Wording must stay identical across every wave — any change breaks comparability with your historical trend and with external benchmarks.

1.
On a scale of 0 to 10, how likely are you to recommend [organization] as a place to work?
QSET-004 — eNPS, adapted from Net Promoter Score methodology (Reichheld & Bain, 2003; HBR)
This is the verbatim generic eNPS question. Promoters score 9–10, Passives 7–8, Detractors 0–6. eNPS = %Promoters − %Detractors. Lock this exact wording for at least 12 months.

Why Behind the Score

Open textFree text

The open-ended follow-ups that turn the single number into a usable signal. One question surfaces the driver; the second reveals the lever. Run both immediately after the rating item in the same survey.

1.
What is the most important reason for the score you gave?
QSET-004 — eNPS
The eNPS number tells you where loyalty stands; this open-end tells you why. Thematic coding across responses reveals the top drivers — and Detractor language in particular surfaces retention risks before they become resignations.
2.
What is one thing that would raise your score?
QSET-004 — eNPS
A forward-looking prompt that converts the survey into an idea channel. Respondents nominate specific, actionable changes rather than just venting — making the follow-up plan easier to write.

Diagnostic Follow-Up

5-pt LikertStrongly disagree → Strongly agree

Two optional questions to run quarterly alongside the core eNPS when you want to understand the loyalty gap without running a full engagement census. Keep these short — the power of eNPS is its brevity.

1.
I see a clear path to grow my career here.
Growth opportunity is consistently the top theme in Detractor and Passive open-ends. Including a single closed-ended growth item lets you quantify the size of the gap without expanding to a full engagement survey.
2.
The way this organization is run makes me proud to work here.
Pride of association is the emotional driver closest to the recommendation impulse measured by eNPS. A low score here, paired with a low eNPS, signals a culture or leadership issue rather than a compensation issue.

Decision Guide

When Should You Use This Survey?

Match the survey type and cadence to your situation.

⚡

You need a fast, recurring read on employee loyalty without running a full census

Use

Run eNPS quarterly on the standard 3-question set (rating + 2 open-ends)Track the trend line — movement matters more than any single scorePair with pulse if you want driver context alongside the loyalty number

Avoid

Swapping in ad-hoc questions or changing the scale between waves

eNPS earns its value from comparability over time. One wording change invalidates the trend.

🚨

Your eNPS is below 0 or has dropped more than 10 points in a single wave

Use

Run an immediate deep-read on the Detractor open-ends — code for top themesEscalate to an engagement survey or targeted pulse to identify root driversShare findings with leadership within two weeks and commit to visible action

Avoid

Waiting for the next quarterly wave to diagnose the drop

A negative or sharply falling eNPS is an early attrition signal — treating it like routine data delays the response and raises flight risk.

📊

You want to understand why engagement is low, not just that it is

Use

Use eNPS as the triage tool — confirm a problem existsThen run a full engagement survey (Gallup Q12-style drivers) to diagnose the causeRoute back to the enps page after the census to track whether the intervention moved the score

Avoid

Relying on eNPS alone to explain complex engagement dynamics

eNPS tells you the outcome (would they recommend you); the engagement census tells you the drivers (why they feel that way).

🧭

You are benchmarking your eNPS against an industry figure

Use

Identify which dataset matches your workforce (Perceptyx for global/large-enterprise; QuestionPro for US/mid-market)Use your industry segment benchmark, not the cross-industry averageAlways label the source and the year when reporting externally

Avoid

Mixing Perceptyx and QuestionPro averages in the same slide deck

The two datasets diverge significantly (12–14 vs 32) due to different sample compositions. Mixing them without labeling produces a meaningless 'average of averages.'

Benchmarks

What "Good" Looks Like

Scores only mean something against a benchmark. Here are the numbers worth measuring against.

12–14

Global eNPS average (Perceptyx, 20M+ employees, 2024). This broader, global sample skews lower than US-only datasets — use it when your workforce is multinational.

Perceptyx Benchmark Database, 2024(vendor-reported)

Global eNPS average reported by QuestionPro for 2025 (up from 25 in 2024). This dataset skews toward US-based, EX-active organizations — benchmark with it if your peers are US mid-market.

QuestionPro, 2025 eNPS Benchmark Report(vendor-reported)

~66 (QuestionPro) / mid-to-high 20s (Perceptyx)

IT sector eNPS — the highest-scoring industry in both datasets. If you are in tech, your baseline expectation should be meaningfully higher than the cross-industry average.

Zonka Feedback synthesis of QuestionPro 2025 + Perceptyx + SurveyMonkey (STAT-006)(vendor-reported)

~11

Government sector eNPS (QuestionPro 2025 synthesis). If you are in the public sector, the cross-industry benchmark overstates what is achievable in the short term — use the government-sector figure as your baseline.

Zonka Feedback synthesis of QuestionPro 2025 + Perceptyx + SurveyMonkey (STAT-006)(vendor-reported)

~30 (startups 0–250) vs ~9 (enterprises 5,001+)

eNPS by company size. Smaller organizations consistently score higher — closer working relationships and fewer hierarchical layers drive more advocacy. Always benchmark within your size band.

Zonka Feedback synthesis of QuestionPro 2025 + Perceptyx + SurveyMonkey (STAT-006)(vendor-reported)

Do It Right

Survey Design Best Practices

The methodology that separates a survey people answer honestly from one they ignore.

Lock the wording and scale for at least 12 months

The standard question wording is load-bearing. Variants — adding the word 'genuinely,' switching to 'extremely,' or moving to a 1–10 scale — produce measurably different scores and break comparability with your historical trend and with any external benchmark. Commit to the exact wording and the 0–10 scale before your first wave, and do not change it mid-program.

METH-010 — eNPS methodology integrity (Sopact; Lattice; Reichheld/Bain guidance)

Never average individual scores — aggregate to group level

eNPS is not a mean. Averaging 10 individual scores of 7 gives you '7,' not an eNPS of anything. The formula requires grouping: count everyone who scored 9–10 as a Promoter, 0–6 as a Detractor, and compute percentages from the full headcount. Running the wrong math produces a number that looks reasonable but cannot be compared to any published benchmark.

METH-010 — eNPS methodology integrity

Benchmark three ways: internal trend first, then external

Your most actionable read is your own trend line — is the score moving up or down quarter-over-quarter? External benchmarks (Perceptyx 12–14; QuestionPro 32) provide industry context, but they diverge enough that you should always name your source and never mix datasets in the same sentence. Internal trend is more actionable than any single external number.

METH-009 — benchmarking method (Culture Amp; HeartCount; Perceptyx)

Do not tie eNPS to manager pay or performance reviews

Linking the score to compensation creates a powerful incentive to coach employees on their responses rather than improve the experience. Fred Reichheld — who developed the original Net Promoter Score — explicitly warned against tying NPS to compensation for exactly this reason. If leaders know their bonus depends on the score, the score stops measuring what you think it measures.

METH-010 — eNPS methodology integrity; Reichheld & Bain, 2003

Suppress results for groups smaller than n=5

In a team of four, a single Detractor response represents 25% of headcount — and anyone who knows the team can potentially identify the respondent. Apply the same n≥5 reporting threshold here that governs your wider survey program. Never report a subgroup score when doing so would re-identify individuals.

METH-010 — eNPS integrity; 15Five Rule of 5 (STAT-010)

Avoid These

Survey Mistakes That Wreck Your Data

Changing the question wording between waves

The most common eNPS mistake is 'improving' the question — adding a qualifier, tweaking the scale, or making it more company-specific — between surveys. Because eNPS is a relative metric, any wording change makes it impossible to compare Wave 2 to Wave 1 or to any published benchmark.

Instead: Finalize your exact wording before launch and treat it as locked for a minimum of 12 months. Document the version you run.

Averaging individual scores instead of computing the real formula

Many tools and spreadsheets compute the 'average' of 0–10 responses and call it eNPS. This is mathematically wrong. An average of 6.8 is not an eNPS — it has no relationship to the −100 to +100 scale and cannot be benchmarked against any published figure.

Instead: Group responses into Promoters (9–10), Passives (7–8), and Detractors (0–6). Compute %Promoters and %Detractors from total headcount, then subtract.

Tying eNPS to manager performance ratings or bonuses

Once managers know their evaluation depends on the score, they start coaching responses rather than improving the experience. Reichheld and Bain warned against this directly when they introduced NPS in 2003. The score becomes a lagging indicator of coaching pressure rather than a leading indicator of loyalty.

Instead: Use eNPS as a diagnostic signal for leadership development — not as an accountability metric tied to variable pay.

Reporting eNPS for groups smaller than five people

In a five-person team, one Detractor represents 20% of headcount. Anyone familiar with the team can often identify who gave the low score, which destroys the psychological safety that honest responses require.

Instead: Apply an n≥5 minimum reporting threshold across all subgroup breakouts. Suppress scores for any group below that threshold and communicate the reason transparently.

Collecting the data and never closing the loop

Running a quarterly eNPS survey that produces no visible action teaches employees that the survey is theater. Perceptyx data shows that only 51% of employees report actual improvements resulted from survey feedback — the gap between collecting data and acting on it is the primary reason participation rates decline over time.

Instead: Publish a 'you said, we heard, here is what we are doing' summary within two weeks of every wave. Even if the action is small, the communication is essential.

Treating eNPS as a substitute for a full engagement survey

eNPS tells you whether employees would recommend you. It does not tell you why, which drivers are at risk, or how to intervene. Using eNPS as the only listening mechanism means you will know a problem exists weeks before you have any idea how to fix it.

Instead: Pair quarterly eNPS with annual engagement surveys and quarterly open-ends. Use eNPS as the early-warning triage tool; use the engagement census to diagnose and act.

Sources & Licensing

Where These Questions Come From

Validated instruments have owners. Here's what's adapted from what — and how to use each one without stepping on a license.

eNPS / Net Promoter Score

eNPS adapts the Net Promoter Score® methodology developed by Fred Reichheld and Bain & Company (2003). The eNPS question wording is generic and freely reproducible. 'Net Promoter Score' and 'NPS' are registered trademarks of Reichheld, Bain & Company, and Satmetrix — attribute the methodology when publishing results or descriptions of the method.

Source: Fred Reichheld, 'The One Number You Need to Grow,' Harvard Business Review, December 2003

Ready to Send

Launch & Follow-Up Templates

The invite, the reminder, and the results share-back — the messages that drive response rates.

eNPS Survey Invite — Email

Subject: One question. Two minutes. Your opinion matters. Hi [First Name], Every quarter we ask one question to understand how employees feel about working here. This wave closes on [Date]. The survey is three questions long and takes under two minutes. [Link] Results are reported at the team level only (groups of five or more), so individual responses are never shared. Your honest answer is what makes the data useful. Thank you, [Sender Name] [Title]

Send from a senior leader or People team — not from an automated platform address. Response rates are higher when the sponsor is visible.

Frequently Asked Questions

eNPS stands for Employee Net Promoter Score. It adapts the Net Promoter Score methodology — originally developed by Fred Reichheld at Bain & Company and published in the Harvard Business Review in December 2003 — to measure employee loyalty rather than customer loyalty. The 'e' prefix is informal and widely used; 'NPS®' and 'Net Promoter Score®' are registered trademarks held by Reichheld, Bain, and Satmetrix.

Keep Exploring

Related Survey Guides

Employee Surveys

What Questions Should Be on an Employee Engagement Survey?

34 employee engagement survey questions grouped by validated driver — Q12-based, UWES-anchored, with the right scale and benchmarks. Copy-ready for your annual census.

Read guide

Employee Surveys

What Questions Should Be on an Employee Morale Survey?

Get 26 validated morale and intent-to-stay questions, flight-risk benchmarks, and copy-ready templates — catch turnover risk before it becomes an exit interview.

Read guide

Employee Surveys

What Is an Employee Pulse Survey (and What to Ask)?

Learn what employee pulse surveys are, when to run weekly, monthly, or quarterly check-ins, and get copy-ready question sets with benchmarks and methodology.

Read guide

Browse All Survey Guides

A Survey Only Helps If You Act On It

The fastest way to tank your next response rate is to collect feedback and do nothing. Actify turns survey findings into action — recognition, engagement activities, and wellbeing benefits employees actually choose.

Book a 20-min Demo See How It Works

No credit card required. 15-minute setup.