eNPS is the employee version of NPS. The score is the easy part. The verbatim names the resignation two quarters before the letter lands.
eNPS is the employee version of Net Promoter Score: "How likely are you to recommend [company] as a place to work?" on a 0-10 scale, with an open-ended follow-up. Most HR teams treat the verbatim as too sensitive to act on — feedback goes anonymous, gets exported, summarized into engagement themes, and the original wording is discarded. Sopact reads every eNPS verbatim on arrival, against the same employee's prior wave — so the resignation signal that was named two quarters ago is visible while HR can still act on it.
By Unmesh Sheth · Founder & CEO, Sopact · Updated May 2026
eNPS (Employee Net Promoter Score) is the employee version of the customer NPS metric. The question: "How likely are you to recommend [company] as a place to work?" on a 0-to-10 scale, with an open-ended follow-up. Same math as customer NPS — percentage of promoters (9-10) minus percentage of detractors (0-6). The score lives on a scale from -100 to 100.
The rating question on a 0-10 scale plus an open-ended follow-up (usually "what is the primary reason for your score?"). Same instrument as customer NPS, asked of employees instead of customers.
Score 9-10: promoter. Score 7-8: passive (excluded). Score 0-6: detractor. Percentage promoters minus percentage detractors. Result lives on the -100 to 100 scale. Negative eNPS is normal in some industries; the math allows it.
An employee whose eNPS dropped from 9 to 4 last quarter and wrote about a specific manager, workload, or career gap is the early resignation signal — usually visible two quarters before the letter arrives. The verbatim is the signal; the score is the trigger.
eNPS shares the same formula as customer NPS, with one wording change. The rating question targets the employer rather than the product; everything else — scale, buckets, calculation — is identical.
"On a scale of 0 to 10, how likely are you to recommend [company] as a place to work?"
eNPS = % Promoters − % Detractors
Promoters 9-10. Passives 7-8 (excluded). Detractors 0-6. Two hundred employees responding with 80 promoters, 70 passives, 50 detractors produces an eNPS of 40 percent minus 25 percent, or 15. The arithmetic is identical to customer NPS; the audience is what changes.
Keep the eNPS rating wording standard — "as a place to work" is the canonical phrasing. Change the company name; do not change the recommendation verb, the scale, or the recipient implication. Benchmark comparability depends on consistent wording, the same way it does for customer NPS.
For ten years HR teams treated eNPS the way they treated other employee-feedback metrics. The score moved each quarter; the team reported it to leadership; the verbatim feedback was summarized into engagement themes and the original wording was discarded. The reasoning was honest — employee verbatims feel sensitive, individual comments feel like surveillance, anonymity feels respectful.
The cost of that reasoning was the resignation signal. Employees who eventually resigned typically named the reason in their eNPS comments one to three quarters before the resignation letter. The specific manager. The specific workload pattern. The career-development gap. The mission misalignment. The data was there. The team was not reading it.
The reading bottleneck is gone. A language model classifies a thousand eNPS verbatims against the team's own codebook in seconds — without anyone in HR having to read raw individual comments where that feels inappropriate. The workflow can be designed to preserve dignity (the eNPS owner sees themes and trajectories; the verbatim text is read only when an intervention is warranted) and still surface the resignation signal that the old summarize-and-discard approach hid. The verbatim can be read responsibly. The choice is no longer between dignity and signal.
More actionable than an engagement-score percentile. More actionable than an exit interview after the resignation. More actionable than a generic "we are listening" memo. The employee named the reason — in their own words — while they were still on the payroll. The team that reads the verbatim catches the resignation while it is still preventable.
This is the same locked argument that anchors /use-case/nps-analysis — expressed here in workforce vocabulary. The pillar covers analysis broadly; this page applies it to employee feedback.
A single employee named "E-04812" in the workforce-feedback system. Their eNPS score and verbatim across five waves on the same persistent employee ID. The aggregate eNPS for the company never moved much. The trajectory on one record told the story two quarters early.
This is not a dashboard chart. It is one employee's record — the only unit at which eNPS analysis can name a resignation in time to prevent it.
eNPS and customer NPS are mechanically identical. The differences sit in the audience, the cadence, the channel, and the sensitivity of the verbatim. Everything else is the same.
Both metrics use the same instrument, the same math, and the same locked thesis (the verbatim is where the work happens). The difference that matters most is the sensitivity of the employee verbatim — and the design challenge is reading it with appropriate dignity protocols rather than discarding it. See /use-case/nps-vs-csat for the broader metric-family comparison.
Each of these is common in mature workforce-feedback programs. Each treats eNPS as something it is not.
The eNPS survey is anonymous; every wave is a fresh respondent set; the converted detractor (was a 9 last quarter, now a 4) cannot be reached because HR does not know which employee they were. Pure anonymity is honest about respect and lossy about signal. Identified-with-rigorous-privacy is the better trade.
The team replaces a fifty-question annual engagement survey with quarterly eNPS and calls it modernization. eNPS measures one thing well — willingness to recommend the employer. It is not a substitute for a deep engagement survey; the two complement each other. Most mature programs run both.
The leadership dashboard shows the company eNPS, the trend line, and the comparison to industry benchmark. It does not show which employees moved, in which direction, on which teams. The aggregate hides the resignation signal it was supposed to surface. Report the trajectory per team and the named-detractor list as well.
Verbatims get classified into themes ("manager," "growth," "compensation") and the original wording is discarded or sealed. The themes become the board report. The wording carries the specific failure mode — which manager, which growth gap, which compensation issue — and the themes do not. Read the verbatim, classify it, and keep the wording on the record.
eNPS lives in one HR system; resignations and exits live in another. The team never goes back to see what the resigned employee's last two eNPS verbatims said. Without that loop, the program cannot improve. Every resignation is an opportunity to read back the eNPS history and confirm whether the signal was there — or whether the survey design failed to capture it.
An eNPS program with a persistent employee ID across quarters is a longitudinal study. The discipline that makes a research-grade longitudinal study work is the same discipline that makes eNPS catch the resignation early.
A research longitudinal study tracks the same units (people, schools, programs) across waves on a persistent ID. The discipline is well established — same instrument, same cadence, attrition planned for, every wave readable against the prior wave on the same record. The most cited example is the Dunedin Study, which has followed 1,037 New Zealanders born in 1972 across more than five decades.
An eNPS program is the same discipline at a different cadence. Same employees (where the persistent ID survives), same rating question, same follow-up, every quarter. The Q3 verbatim reads against the Q2 verbatim against the Q1 verbatim — on one record. That is what makes the resignation signal visible. Without the persistent ID, every wave is a stranger; with it, every wave is a chapter in the same employee's story with the company.
The bridge runs both ways. The longitudinal-design pillar explains why same-unit-across-waves is the load-bearing methodology. This page explains why the same methodology applied to employee feedback catches resignations early. They are the same argument, told in two vocabularies.
The methodology that makes eNPS work as a resignation-prediction instrument — same employee, same instrument, every quarter, on one record. The discipline behind the design.
The methodology, the closed-loop workflow, the detractor handling all apply with employees substituted for customers. Three doors into the same workflow, expressed in customer vocabulary.
What NPS analysis means in 2026 — the methodology that applies to both customer and employee feedback.
Read the pillar →The workforce sibling — same instrument, different audience, same locked thesis. The verbatim names the resignation.
This pageThe closed-loop workflow that applies to either customer or employee feedback — the same five stages, applied to the eNPS comment.
Read the hub →Every other NPS cluster page applies to eNPS with the audience substituted. The detractor workflow handles a detractor employee; the survey-questions page covers eNPS question design; the benchmarks page covers eNPS benchmarks. One cluster, two audiences.
Same eNPS instrument, three different workforce programs. In each case the verbatim is the unit that earns the action; the score is the trigger.
A 500-person SaaS company runs quarterly eNPS with identified responses and a per-team breakdown. The Q3 verbatim from a senior engineer named a specific manager pattern; HR routed the verbatim (without the engineer's name attached) to the engineering VP. By Q5 the manager pattern was addressed and the engineer's score recovered. The resignation that would have arrived in Q6 did not.
A two-year leadership development program runs eNPS at intake, mid-cycle, and exit. Each participant's three responses sit on one record alongside their application essay. The team can see which participants disengaged mid-cycle and what their verbatim said. The next cohort's curriculum is designed on the disengagement signals from the prior one.
A mid-size nonprofit runs eNPS quarterly. The team is mission-aligned but burnout-prone; resignations cluster in the second year. The eNPS verbatim from Q4 of year one consistently names the workload-vs-mission gap. The team can intervene with workload redesign before the second-year resignation curve hits.
Your scores, your verbatims, your employees on a persistent ID. Sixty minutes. No demo accounts.
eNPS stands for Employee Net Promoter Score. It is the employee version of the customer NPS metric, asking "How likely are you to recommend [company] as a place to work?" on a 0-10 scale, with an open-ended follow-up. Same math as customer NPS — percentage of promoters (9-10) minus percentage of detractors (0-6). The score lives on a scale from -100 to 100.
eNPS stands for Employee Net Promoter Score. It is a survey-based metric that measures how likely employees are to recommend their employer as a place to work. The metric was adapted from the Bain-developed customer NPS in the late 2000s and has become the most widely used single-number employee-feedback measure.
eNPS = percentage of promoters (employees who scored 9 or 10) minus percentage of detractors (employees who scored 0 through 6). Passives (7-8) are excluded from the formula. The result is a number from -100 to 100. For an eNPS wave of 200 employees with 80 promoters, 70 passives, and 50 detractors: 40 percent promoters minus 25 percent detractors equals an eNPS of 15.
There is no universal good eNPS. Typical eNPS medians run 0-30; top-quartile employers reach 30-50; leading employers reach 60+. The score varies by industry, company size, and methodology. The more honest measure is whether eNPS is improving on the same employee base across quarters, and whether the verbatim is being read on arrival — the same discipline that matters for customer NPS.
Same instrument, different audience. NPS asks customers if they would recommend the company; eNPS asks employees if they would recommend the company as a place to work. Same 0-10 scale, same formula, same follow-up open-ended question. The differences are in the cadence (eNPS often quarterly or semi-annual), the channel (internal HR systems versus customer surveys), and the sensitivity of the verbatim (employee feedback is often anonymized in ways that customer feedback is not).
The eNPS verbatim is where the resignation signal lives. The score moves; the verbatim explains why. An employee who scored 5 with a verbatim about a specific manager, a specific workload pattern, a specific career-development gap is a resignation in the making — usually two quarters before HR sees the resignation letter. Most HR teams treat eNPS verbatims as too sensitive to act on. That is the most common failure mode in workforce-feedback programs.
Anonymous eNPS feels respectful and produces unactionable data. Without a persistent employee ID across waves, every response is a stranger; the trajectory of a specific employee is invisible; the converted detractor (was a 9 last quarter, now a 4) cannot be reached because HR cannot tell which employee they were. The strongest workforce programs use identified eNPS with rigorous privacy protocols around who can see which verbatim. Anonymity is the easy answer; identified-with-care is the right one.
Quarterly is the most common cadence for relational eNPS. Some teams run monthly pulse eNPS in addition. Annual is too infrequent — the trajectory cannot be seen wave-over-wave. The cadence matters less than the persistent employee ID across waves — a quarterly eNPS with no employee continuity is wave-by-wave reporting, not measurement.
The default follow-up is "What is the primary reason for your score?" Most eNPS surveys use this wording. Better variants: "What would have made this a 10?" (best for detractors), "What should we keep doing?" (best for promoters), "What is one thing we could change about working here?" (action-oriented, broad). Branching the follow-up by score range is the single most-skipped survey-design move in employee feedback programs.
The score alone is a weak predictor; the verbatim is a strong one. Employees who eventually resigned typically named the reason in their eNPS comments one to three quarters before the resignation — specific manager issues, workload, career-development gaps, mission misalignment. The signal is in the wording, not in the score. A program that reads every eNPS verbatim on arrival, against the same employee's prior wave, sees the resignation signal before HR sees the letter.
Engagement surveys typically include dozens of questions and produce composite scores. eNPS is two questions and produces one number plus a verbatim. They measure different things — engagement covers the full job experience; eNPS covers willingness to recommend. Most mature workforce-feedback programs run both, with eNPS quarterly and a deeper engagement survey annually. The mistake is treating eNPS as a substitute for engagement, or engagement as a substitute for the per-employee verbatim that eNPS produces.
eNPS is the workforce sibling of customer NPS — same instrument, different audience, same locked thesis: the score is the easy part; the verbatim is where the work happens. The methodology pillar (/use-case/nps-analysis), the closed-loop workflow (/use-case/nps-feedback), and the detractor handling (/use-case/nps-detractor) all apply with employees substituted for customers. Run with a persistent employee ID across quarters, eNPS becomes a longitudinal signal (/use-case/longitudinal-design) — the same employee's trajectory through the relationship.
eNPS is the workforce sibling of customer NPS. The cluster covers the methodology, the closed-loop workflow, and the time axis — same argument, customer vocabulary.
The pillar above this page — the methodology that applies to both customer and employee feedback. The locked thesis behind the cluster.
The detractor handling workflow — the four sub-types, the five-step follow-up. Applies to detractor employees just as it does to detractor customers.
The methodology that makes eNPS work across quarters — same employee, same instrument, every wave, on one record. The discipline behind the design.
Your eNPS scores, your verbatims, your employees on a persistent ID. Sixty minutes. We classify each verbatim on arrival against your codebook, surface the converted detractors and named patterns, and walk through what reading-in-context would have caught about the last quarter. No demo accounts. No slideware. Your own workforce data, read live with privacy protocols intact.
No slideware. No demo accounts. Your own workforce data, read live.
