eNPS asks how likely someone is to recommend your organisation as a place to work, on a scale of 0 to 10, and subtracts the share of detractors from the share of promoters. It gives you one number between −100 and +100 that you can watch over time.
On its own that number tells you almost nothing about why. So the free-text question next to it is the one that matters, and reading those answers honestly is the work. Timbre does that reading: themes across the responses, how widespread each is, and representative quotes — never one attributed to anybody.
Set a target and watch the distance to it
A score with nothing to compare it against only answers "did it move". You can set the eNPS you are aiming for, with your own label — "Board goal, FY26" — and it is drawn on every eNPS chart with the distance to it shown beside the current score.
If you have a benchmark you trust from elsewhere, you can shade it behind the trend and name its source. We do not supply industry benchmark numbers: we have not licensed any, and a plausible invented average is a number somebody would plan against.
Recurring, without becoming wallpaper
Pulse surveys can run on a schedule — monthly, quarterly, whatever cadence you can actually act between. Reminders are automatic, and the rotating templates stop the third one reading exactly like the first.
Response rates fall when people conclude nothing changes. The reports are built to produce something you can act on, and the AI culture report names what to do rather than only what happened.
What you see
The eNPS trend with promoters, passives and detractors broken out; department comparison; per-question movement over time; and an AI culture report that narrates the movement, including a decline, candidly.
Questions people ask
What counts as a good eNPS?
Anything positive means more promoters than detractors, and above +20 is generally considered strong — but the comparison that matters is against your own last score, not a published average. Benchmarks vary so much by industry, country and how the question was asked that a number lifted from a blog post is worse than no number. Set a target you can defend and track the distance to it.
How often should we run it?
Quarterly suits most companies under 100 people. Monthly is defensible if you are actively changing things and want to see whether it landed; more often than that and you are asking people to repeat themselves before anything has had time to change, which is how response rates collapse.
Is it really anonymous?
Responses are never displayed attributed to a person, and written comments stay hidden until a minimum number of people have answered so that a comment cannot be traced back by process of elimination. The thresholds are published rather than described vaguely, so employees can check the claim.
Can we ask more than the one question?
Yes — most companies pair eNPS with two or three others and one free-text prompt. Keeping it short is the main thing: a five-minute survey that people finish beats a thorough one they abandon.
$49/month
Up to 50 people · every feature · price locked for 12 months