Burnout is the leading indicator. Attrition is the receipt.

The short answer

Nobody resigns on the day they decide to. There is a gap, usually months, between the decision forming and the message landing in your inbox. Everything useful you could have done lives inside that gap, and every number most companies track lives outside it.

Burnout is not the only thing that fills that gap. But it is the one that accumulates slowly, shows up in how people answer ordinary questions, and is measurable without reading anybody’s messages. That combination is rare, and it is why it is worth measuring even if you do not care about wellbeing for its own sake.

Attrition is a receipt

Your attrition rate went from 11% to 16%. Good. What do you do on Monday?

The honest answer is that you cannot do anything about the number itself, because every person inside it has gone. You can respond to what it implies, slowly, and find out in two quarters whether the response worked. That is a long feedback loop for an expensive problem.

Gallup, in research published in March 2019, found that 52% of people who left voluntarily said their manager or organisation could have done something to keep them. Not 52% who were unhappy. 52% who believed the outcome was avoidable. In the same work, 51% said that in the three months before they left, no manager or leader had spoken to them about their satisfaction or their future.

Put those two together and you get the shape of the problem. The thing that would have worked was a conversation. The reason it did not happen was that nobody knew it was needed.

The route from tired to gone

The WHO describes burnout through three things: exhaustion, a growing mental distance or cynicism about the job, and a falling sense of being effective at it. It classifies it as an occupational phenomenon arising from chronic workplace stress that has not been successfully managed, not as a medical condition.

Read that list again as a sequence rather than a definition, and it describes a resignation forming.

1. Exhaustion. The weeks get heavier. Recovery between them stops happening. Nothing visible changes at work, because people absorb it.
2. Distance. Caring becomes expensive, so it gets rationed. The person is still delivering, and still saying “fine”, but has quietly stopped arguing in design reviews.
3. Reduced efficacy. They no longer feel good at the job. At this point the story changes from “this place is hard” to “I am not right for this place”, which is a story you leave over.
4. A conversation elsewhere. A recruiter, a friend, an old colleague. The offer is not what caused this. It is what made it actionable.

By the time anyone at the company notices, it is usually step four. The first three were visible for months, to anyone measuring.

What a leading indicator has to do

Three tests. Most of the things companies already run fail at least one.

That last row is the only one that passes all three, and it passes the third conditionally. Get the anonymity wrong and you have built the fourth row with better branding, which is the subject of burnout detection versus employee monitoring.

How you measure it without surveillance

No single signal is worth much. A bad mood on a Tuesday is a bad Tuesday. The useful thing is a few honest signals, combined, over weeks.

Wanderfly combines four per department: stress from the monthly PSS-10 scale, meeting workload read from calendar timing, daily mood, and engagement, meaning whether people are still answering at all, which is often the first thing to go. It recomputes over the last 28 days and produces a score from 0 to 100 with a confidence level, plus a forecast at 14, 30 and 60 days.

What it will not do is tell you who. No leader can see an individual’s mood, stress score, conversations or expert bookings, and no group smaller than five people gets a view at all. That is a deliberate limit, and it is also what makes the data worth having: the moment people believe an answer is attributable, they stop giving you honest ones, and then you are measuring politeness.

Where this argument stops

We sell a burnout detection product, so it is worth being plain about what this page is not claiming.

• Burnout does not explain every exit. People leave for money, for a title, for a city, for a partner’s job, for a founder they no longer believe in, and because a better company called. A rested person can still take a 40% raise.
• A high score is not a prediction about a person. It is a statement about a team’s conditions over four weeks. Teams with high scores do not always lose people, and teams with low ones sometimes do.
• Measuring it changes nothing by itself. A dashboard nobody acts on is a more expensive way of being surprised. The signal is only worth the intervention attached to it.
• Some attrition is fine. A company with no voluntary exits is usually not being honest with itself about performance.

What we do claim is narrower and, we think, defensible: of the reasons people leave, burnout is among the most common, the most expensive, and almost uniquely the most visible in advance, if anybody is looking.

Where to start

If you have never measured anything beyond the exit interview, do not start by buying something. Start by working out your real number, split voluntary from involuntary, and see which department it is concentrated in. The attrition rate formula and calculator will take you twenty minutes.

Then pick one team and start asking, often and anonymously. If you want the version with a method behind it, that is what Wanderfly does, and how it fits a retention plan. If you would rather do it with a spreadsheet and a recurring calendar invite, do that. The measurement matters more than the tool.

Questions people ask