How to Calculate Employee Net Promoter Score (eNPS): The Complete Guide for 2026
Last Updated June 24, 2026
The eNPS calculation is one of the simplest formulas in people analytics. Ask employees how likely they are to recommend the company as a place to work on a zero-to-ten scale, classify each response as a promoter, passive, or detractor, and subtract the detractor percentage from the promoter percentage. The math takes thirty seconds once you have the responses. What takes judgment is understanding what the number means, what affects its accuracy, how to break it down by team, and what to do with it once you have it.
This guide walks through the complete eNPS calculation — the formula, a worked example, the common mistakes that produce misleading scores, how to calculate team-level eNPS alongside the organization-wide number, and how to interpret the result in a way that produces genuine insight rather than just a metric to report at the next all-hands.
The eNPS Formula
eNPS = % Promoters − % Detractors
That is the entire formula. The passives — employees who answered seven or eight — are excluded from the calculation entirely. They are neither actively advocating nor actively discouraging, and their scores are not added or subtracted from the result. They matter for understanding your score and for thinking about retention strategy, but they play no role in the arithmetic.
The three categories are defined by the response to the core question — "On a scale of 0 to 10, how likely are you to recommend this company as a place to work to a friend or colleague?" — as follows:
Promoters: employees who answered 9 or 10. These are the employees most likely to actively recommend the organization in conversations you're not present for — to candidates who ask what it's like to work there, to friends considering a job change, to professional contacts deciding between employers.
Passives: employees who answered 7 or 8. These employees are not actively discouraging others from joining, but they are not actively advocating either. They are the most movable segment — a single significant positive or negative experience can shift them toward promoter or detractor status.
Detractors: employees who answered 0 through 6. These are employees who are either actively discouraging others from joining or would do so if asked directly. A detractor who is asked by a friend whether your company is a good place to work will say no — or at minimum will say something that does not help your recruiting. Detractors are also significantly more likely to be planning to leave than promoters or passives.
A Worked Example
Suppose you send an eNPS survey to 200 employees and receive 160 responses — an 80% response rate. The responses break down as follows:
Scores of 9 or 10: 80 responses
Scores of 7 or 8: 48 responses
Scores of 0 through 6: 32 responses
Step 1: Calculate the percentage of promoters.
80 promoters ÷ 160 total responses = 50%
Step 2: Calculate the percentage of detractors.
32 detractors ÷ 160 total responses = 20%
Step 3: Subtract detractor percentage from promoter percentage.
50% − 20% = +30
Your eNPS is +30.
The passives — 48 responses, or 30% of the total — do not appear in the formula. They are accounted for implicitly: because promoters are 50% and detractors are 20%, the remaining 30% are passives by subtraction. But they are not subtracted from or added to the score. The only numbers that matter for the calculation are the promoter percentage and the detractor percentage.
The Range of Possible eNPS Scores
eNPS scores range from −100 to +100. A score of −100 would mean that every single respondent is a detractor — every employee answered zero through six. A score of +100 would mean that every single respondent is a promoter — every employee answered nine or ten. Neither extreme occurs in practice, but understanding the range helps contextualize where any given score sits.
A score of 0 means that exactly as many detractors as promoters responded — the organization is perfectly split between those who would advocate for it and those who would discourage others from joining, with the passives making up whatever is left. A score below zero means detractors outnumber promoters. A score above zero means promoters outnumber detractors.
Most organizations score somewhere between −20 and +60, with meaningful variation by industry, company size, and the economic conditions at the time of the survey. Technology companies tend to score higher than retail or healthcare. Companies in periods of growth and stability tend to score higher than those undergoing restructuring or leadership transitions. Benchmarks vary significantly by source and population, which is one of the reasons your own trend over time is more useful than any external benchmark for most management purposes.
What Counts as a Good eNPS Score
Scores above 0 indicate that promoters outnumber detractors — a basic positive signal. Scores above +20 are generally considered good across industries. Scores above +40 are considered excellent. Scores above +60 are exceptional and are typically found only in organizations with unusually strong cultures, high-quality management, and compensation well above market.
Scores below 0 are a meaningful warning signal. An organization where detractors outnumber promoters has more employees actively discouraging others from joining than actively recommending it — which damages recruiting, accelerates attrition, and reflects conditions that, if unaddressed, tend to compound over time rather than self-correct.
The most important benchmark for most organizations is not the industry average but the organization's own score over time. A score of +15 that has risen from +5 over three quarters signals improving conditions. A score of +35 that has fallen from +50 over the same period signals deteriorating conditions that the still-positive absolute score may obscure. The direction and rate of change tell you whether what you're doing is working. The absolute score tells you where you are. Both are necessary for a complete picture.
How to Calculate eNPS by Team
The organization-wide eNPS score is useful as a headline metric and a trend indicator. The team-level eNPS breakdown is where the most actionable data lives. An organization-wide score of +30 that is the average of a +60 in engineering and a +5 in customer support is a very different management situation from one where all teams score near +30. The variation between teams is almost always larger than the variation in the organizational average over time, and it is almost always where the most specific, most addressable retention risk is concentrated.
Calculating team-level eNPS uses exactly the same formula applied to each team's responses separately. Using the same example as above, suppose the 160 responses break down by department as follows:
Engineering (40 responses): 26 promoters, 10 passives, 4 detractors
Sales (50 responses): 22 promoters, 18 passives, 10 detractors
Support (38 responses): 14 promoters, 12 passives, 12 detractors
Operations (32 responses): 18 promoters, 8 passives, 6 detractors
Engineering eNPS: (26 ÷ 40) − (4 ÷ 40) = 65% − 10% = +55
Sales eNPS: (22 ÷ 50) − (10 ÷ 50) = 44% − 20% = +24
Support eNPS: (14 ÷ 38) − (12 ÷ 38) = 37% − 32% = +5
Operations eNPS: (18 ÷ 32) − (6 ÷ 32) = 56% − 19% = +37
The organization-wide score of +30 is the weighted average of these four team scores — but the team-level breakdown reveals that support at +5 is a very different situation from engineering at +55, and that the organization-wide average obscures a retention risk that is concentrated in one specific team. The manager responsible for the support team needs a very different conversation than the manager responsible for engineering, and neither of those conversations would be prompted by a report that shows only the organization-wide average.
One important note on team-level eNPS: minimum respondent pool size. Calculating eNPS for a team of four people produces a number that is both statistically unreliable — a single response change can move the score by twenty-five points — and potentially de-anonymizing, because the respondent pool is small enough that specific scores may be traceable to individuals. Most practitioners set a minimum threshold of eight to ten responses before reporting team-level eNPS as a meaningful number. For teams below that threshold, aggregate the data with adjacent teams or report only the organizational average without team-level breakdowns.
The Denominator Question: Total Employees or Total Respondents
One of the most common sources of confusion in eNPS calculation is what to use as the denominator: the total number of employees surveyed, or the total number who responded. The standard practice, and the one that produces comparable scores across survey cycles, is to use total respondents — not total employees surveyed.
The reason is straightforward: non-respondents have not expressed a preference, and including them in the denominator would dilute the score in a way that conflates low response rate with low sentiment. An organization where 40% of employees respond and 70% of respondents are promoters is in a different position from one where 90% respond and 70% of respondents are promoters — but the eNPS formula should produce the same score in both cases, because the question being answered is what the respondents' sentiment is, not what the non-respondents' silence means.
What this means practically is that response rate is a separate metric from eNPS score, and both should be tracked and reported together. A high eNPS from a low response rate is a qualified result — it may be that the employees most likely to be detractors were also the least likely to complete the survey. Tracking response rate by team alongside eNPS by team often reveals that the teams with the lowest response rates are also the teams with the most disengaged employees — which is itself a retention signal worth acting on, separate from the score that was calculated from the responses that arrived.
Common Mistakes in eNPS Calculation
Including passives in the denominator differently than respondents. The denominator for both the promoter percentage and the detractor percentage is total respondents — including passives. Some calculators mistakenly exclude passives from the denominator, which inflates both percentages and produces a score that is not comparable to standard eNPS methodology. Always divide the number of promoters by total respondents and the number of detractors by total respondents, with passives counted in the denominator but not in the numerator of either calculation.
Averaging team scores to get the organizational score. The organization-wide eNPS is calculated from the full respondent pool — all promoters divided by all respondents, minus all detractors divided by all respondents — not by averaging the team-level scores. Averaging team scores would give equal weight to a team of five and a team of fifty, which produces a distorted organizational figure. Always calculate the organizational score from the raw response data, and calculate team scores separately from the same data.
Comparing scores across different response rates without noting the difference. An eNPS of +40 from an 85% response rate is a more reliable signal than an eNPS of +40 from a 35% response rate. Both calculate correctly, but the second figure is based on a respondent pool that may not be representative of the broader employee population. Always report response rate alongside eNPS score, particularly when response rates vary meaningfully between survey cycles or between teams.
Treating a single score as meaningful without a trend. A single eNPS measurement tells you where sentiment is at one point in time. It doesn't tell you whether conditions are improving or deteriorating, which is the most useful thing the metric can tell you. A score of +25 measured for the first time carries very little interpretive weight. A score of +25 measured after three consecutive quarters of decline from +45 carries a great deal. Run eNPS on a consistent cadence and track the trend alongside the absolute score.
Rounding percentages before subtracting. Round the final eNPS score, not the intermediate percentages. If promoters are 47.5% and detractors are 18.3%, the eNPS is 47.5 − 18.3 = 29.2, which rounds to 29. Rounding promoters to 48% and detractors to 18% before subtracting produces 30 — a one-point difference that accumulates into meaningful trend distortion across multiple survey cycles. Do the subtraction first, then round the result.
How Anonymous Surveys Affect eNPS Accuracy
The accuracy of your eNPS score depends entirely on whether employees are answering honestly. Employees who don't trust that their response is genuinely anonymous score conservatively — toward the middle of the scale — rather than recording their actual sentiment. The result is a score that is systematically compressed toward zero: fewer nines and tens from employees who are genuinely satisfied but don't want to seem sycophantic, and fewer low scores from employees who are genuinely dissatisfied but don't want to risk identification.
A compressed eNPS score trends more slowly, responds less clearly to genuine improvements or deteriorations in employee conditions, and understates both the advocacy of your most satisfied employees and the dissatisfaction of your most at-risk ones. It is, in short, a less useful metric — not because the formula is wrong, but because the inputs are wrong.
Technical anonymity — enforced by the survey tool's architecture rather than promised in a policy statement — produces more honest scores because employees can verify that their individual response cannot be traced back to them. The difference in score between a technically anonymous survey and a nominally anonymous one is typically several points in both directions — promoter percentages are higher because genuinely satisfied employees answer honestly, and detractor percentages are also sometimes higher because genuinely dissatisfied employees feel safe saying so. The result is a more volatile score in the right direction — one that actually moves in response to changing conditions — and a more reliable trend signal over time.
How to Interpret Your eNPS Score
Interpretation requires three reference points: the absolute score, the trend, and the driver data. The absolute score tells you where sentiment is. The trend tells you whether conditions are improving or deteriorating. The driver data — the follow-up questions and open-text responses that accompany the eNPS question — tells you why.
An eNPS of +35 that has been stable for four consecutive quarters with no change despite management attention and investment is telling you that the conditions driving the score are structural — compensation competitiveness, growth ceiling, workload — and not addressable through management behavior changes alone. An eNPS of +35 that has risen from +10 over four quarters following a specific management intervention is telling you that the intervention is working and should be sustained and expanded.
The open-text follow-up question — "what is the main reason for your score?" — is where the specific narrative behind the number lives. Aggregate the open-text responses by theme: how many responses mention manager quality, growth opportunity, compensation, workload, culture, or specific organizational events? The themes that appear most frequently among detractor responses are the conditions most urgently driving low scores. The themes that appear most frequently among promoter responses are the conditions worth protecting as the organization grows and changes.
Team-level interpretation follows the same logic applied at a smaller scale: which teams have the lowest scores, which have the steepest negative trends, and what do the open-text responses from those teams say about why. The manager responsible for a team with an eNPS of +5 and declining needs a different conversation than the manager with an eNPS of +55 and stable — and both need a more specific conversation than "your team's engagement score is low" can produce.
Presenting eNPS Results to Leadership
When presenting eNPS results to senior leadership, lead with the trend rather than the absolute score. A score of +30 means more when the room understands that it was +18 six months ago and +8 a year ago than when it's presented as a standalone number. The trend is the evidence that something is working or not working — and leadership decisions about where to invest, which teams need attention, and which managers need support are better made with trend data than with a single-point snapshot.
Present team-level scores alongside the organizational average. The variation between teams is almost always the most actionable part of the data — it identifies where the retention risk is concentrated, which managers are producing the conditions that retain people, and where HR and leadership attention should be directed in the next quarter. An organization-wide score presented without team-level breakdown is a summary without a story.
Connect the score to the open-text themes. The most effective eNPS presentations to leadership include three to five direct quotes from the open-text follow-up — specifically from detractors — that illustrate concretely what is driving the score. Aggregate data tells leadership where the problem is. Direct employee language tells them what it feels like from the inside, which is more persuasive and more memorable than a percentage point.
Calculate eNPS Automatically with FormRoyale
FormRoyale calculates your eNPS score automatically as responses arrive — no spreadsheet, no formula, no export step. Build your eNPS survey in minutes, share a URL with your team, and watch the score update in real time on a dashboard that also shows the promoter and detractor percentages, the response distribution across the zero-to-ten scale, and the open-text follow-up responses alongside the score data. Team-level segmentation shows the score breakdown by department so the variation between teams is immediately visible without any manual calculation.
Flat pricing at $14.50/month covers unlimited eNPS surveys, unlimited questions, and unlimited responses. No per-seat costs, no response caps, no upgrade prompts. One plan, every feature, any team size.
→ Try FormRoyale free for 7 days — no credit card needed
Frequently Asked Questions
What is the eNPS formula?
eNPS = % Promoters − % Detractors. Promoters are employees who answered 9 or 10 on the zero-to-ten recommendation scale. Detractors are employees who answered 0 through 6. Passives — employees who answered 7 or 8 — are included in the denominator when calculating percentages but are not added or subtracted in the final formula. The result is a score between −100 and +100.
Do you include non-respondents in the eNPS calculation?
No. The standard eNPS calculation uses total respondents — not total employees surveyed — as the denominator. Non-respondents have not expressed a preference and including them would conflate response rate with sentiment. Track response rate as a separate metric alongside the eNPS score, and note when response rates are low enough to raise questions about whether the respondent pool is representative of the broader employee population.
How do you calculate eNPS for a small team?
The formula is the same regardless of team size — promoter percentage minus detractor percentage. The practical consideration for small teams is minimum respondent pool size. Calculating eNPS from fewer than eight to ten responses produces a statistically unreliable number — a single response change can move the score by ten to twenty-five points — and may compromise anonymity if the respondent pool is small enough that individual scores are traceable. For teams below the minimum threshold, report results in aggregate with adjacent teams or report only the organizational average without team-level breakdowns.
What is the difference between eNPS and NPS?
The formula is identical — both subtract the detractor percentage from the promoter percentage using the same zero-to-ten scale and the same category boundaries. The difference is the question being asked. NPS asks customers how likely they are to recommend the company's product or service. eNPS asks employees how likely they are to recommend the company as a place to work. The methodology is borrowed from customer experience measurement and adapted for the employee experience context. eNPS scores are not comparable to NPS scores — the populations, the stakes, and the typical score distributions are different — but the calculation and interpretation logic are the same within each context.
How often should you recalculate eNPS?
Quarterly is the most appropriate recalculation cadence for most organizations. Quarterly surveys are frequent enough to catch meaningful sentiment shifts before they become retention problems, and infrequent enough that the question doesn't become so routine that employees answer it without reflection. For organizations going through significant change — restructuring, leadership transitions, major policy shifts — a more frequent cadence of six to eight weeks may be appropriate to track how sentiment is moving through the change period. Annual recalculation is too infrequent: most departure decisions are made within six to twelve months of first experiencing the conditions driving them, and an annual survey catches too many of them after the decision is already made.
Can eNPS scores be negative?
Yes. A negative eNPS score means that detractors outnumber promoters — more employees answered zero through six than nine or ten. A score of −15, for example, means the detractor percentage exceeds the promoter percentage by fifteen points. Negative scores are a meaningful warning signal: an organization where detractors outnumber promoters has more employees actively discouraging others from joining than advocating for it, which damages recruiting, signals elevated attrition risk, and reflects conditions that tend to compound if unaddressed. Negative scores are not uncommon during periods of organizational difficulty — restructuring, leadership transitions, compensation compression — but sustained negative scores across multiple quarters indicate a structural problem that requires direct intervention rather than time alone to resolve.