Burnout is almost always spotted in hindsight. First someone stops taking leave, then they start replying in Slack at eleven at night, then they hand in their notice — and the exit interview reveals it had been building for six months. Every step was visible in data the company already had. Nobody was putting the two halves together.
Wellbeing in DTPulse puts them together: what people say about their workload, next to how much they actually rest. On its own each half lies. Together they produce a signal you can open a conversation with.
Six statements
Wellbeing is a ready-made pulse survey preset. Six statements on a 1–5 agreement scale: current workload, the ability to switch off, energy at the end of the day, recovery during time off, support from the manager, and the norms the team operates by. The wording is fixed — you don't have to invent it and you can't accidentally ruin it with a leading question. The default cadence is quarterly.
One person's six answers collapse into a single index from 0 to 100. Those indices are then averaged by department, by manager and by tenure bracket — and shown only where a slice holds three or more responses. The anonymity mechanics are the same as for every other pulse survey: the response table carries no employee reference, submission time is rounded to the hour, and the three-response threshold is written into the SQL query. A two-person department has no index, not because the interface hides it but because the query doesn't return it.
Every current value is shown next to the previous run and the difference between them. "68" on its own means nothing. "68, down from 74 last quarter" means something.
The other half: actual rest
A subjective index is easy to be reassured by — people answer cheerfully on a form their employer sent. So the department table puts two absence-derived columns next to the index: the average accumulated leave balance, and the average number of rest days taken over the last 12 months. Both are computed from the absence types you have flagged as counting as rest — vacation and days off by default, adjustable to your policy. Only employees with at least a year of tenure count, and the columns appear only when a department has at least three of them.
The interesting reading is where the halves meet. An index of 72 with an average balance of six unused days is a normal team. An index of 72 with thirty-four days of unused leave and eight rest days taken all year is a team telling you everything is fine while not resting.
The card on an employee profile
The right-hand sidebar of an employee page carries a card with three figures: when they last rested, how much leave they have accumulated, and how many rest days they've taken in 12 months. Each is tinted red through amber to green, so the state reads at a glance without arithmetic.
The card is visible to the person themselves, their managers — including heads of departments further up the tree — HR and admins. Colleagues don't see it, and not merely because the interface hides it: the endpoint refuses them.
A manager opening the aggregate results can also narrow them to their own area. That filter honours all four management channels: direct reports, manually assigned reports, departments the person manages manually, and the whole subtree of departments they head. The three-response threshold applies there too — a manager of two people will not see their index.
What the index does not tell you
It isn't a diagnosis. Six statements about balance are a thermometer, not a blood test: they tell you a conversation is worth having, not what it should be about.
It isn't about hours worked. DTPulse doesn't track time at the keyboard and doesn't read calendars — the objective half is built from leave and days off, which is recovery, not load.
It doesn't point at an individual. The index is always a group figure and always gated; the personal card on a profile is built from recorded absences, not from survey answers. No individual answer can be tied back to a person — which is exactly what makes candid answers possible.
Whose time this saves
HR gets a recurring wellbeing measurement without a separate CultureAmp, Lattice or Officevibe subscription, and without shipping employee records into one more external service.
Managers get a defensible reason to book a one-to-one: not "I have a feeling", but "your team's index dropped nine points and the average unused leave balance is twenty-nine days".
Leadership gets departments on one comparable scale and a view of where load is distributed unevenly.
Employees get a way to flag being overloaded without putting their name on it.
Getting started
Check which absence types count as rest in your absence settings. Create a Wellbeing survey on a quarterly cadence — the questions are already written, there is nothing to configure. One quarter later you have a second data point and the index starts to mean something.
Wellbeing is a pulse survey format, so it occupies one active-survey slot: 1 on the free tier, 3 on TEAM, 10 on BUSINESS, 50 on ENTERPRISE. The leave data is already in the portal — nothing extra to set up.
For self-hosted deployments this matters twice over: the most sensitive signal a company collects about its people never leaves your own database, and there is no external engagement vendor in your processing records.