Why employees leave (and the signals you already have)

Key takeaways

• Resignations rarely come out of nowhere. Most are the end point of a slow drift that started months, sometimes years, earlier.
• The common reasons people leave cluster around a stalled path, a poor manager relationship, feeling unheard, and a workload that never recovered.
• You already collect data that shows this drift: engagement scores, development activity, one-on-one patterns and internal mobility.
• The skill isn't collecting more data. It's reading the data you have as a retention signal instead of a compliance box to tick.
• Want to know what that drift is actually costing you? Our turnover cost calculator will put a real number on it.

A manager once told me she was "blindsided" when her top performer resigned. I asked her a few questions. He'd stopped putting his hand up for stretch projects around eight months earlier. His engagement score had slid from an 8 to a 5. He'd cancelled three one-on-ones in a row and rescheduled none of them.

She wasn't blindsided. She just hadn't been reading the signal. Nobody sets out to be surprised by a resignation. But most of us are, more often than we'd like to admit, because we treat leaving as a single event rather than the last step in a process that's been visible for months.

The common reasons people actually leave

I've had this conversation with hundreds of leaders and employees over the years, and the reasons cluster in a pretty small number of places.

They can't see where this goes. Not "there's no promotion this year." More often, nobody has ever had a real conversation with them about their growth, so they assume there isn't one to be had. For more on fixing this specifically, see career pathing that actually retains people.

Their manager let them down. Inconsistency, favouritism, or simply being unsupported through something difficult. Of everything on this list, the manager relationship is the one that shows up most often when I ask people, honestly, what finally tipped them.

They stopped feeling heard. Feedback went into a survey once a year and nothing visibly changed. Eventually people stop bothering to say what they think, and that silence is itself a signal worth watching for.

Something felt unfair. A role that went to someone less capable. Recognition that seemed to follow relationships rather than results. People will tolerate a lot of things. Unfairness compounds faster than almost anything else.

The workload never recovered. A tough quarter is normal. A tough year that becomes the new normal, with no acknowledgement it's happened, tells people their wellbeing isn't actually a priority, whatever the policy documents say.

For the fuller picture of what keeps people versus what pushes them out, the complete guide on employee retention lays out both sides properly.

The signals show up long before the resignation

Here's the part I want leaders to sit with. Every one of the reasons above has an earlier, quieter version. People don't wake up one day and decide to leave. They disengage first, and disengagement leaves a trail.

Engagement scores drift down for that person specifically. Not the team average, which can mask an individual decline entirely. Watch the trend for the people you'd genuinely miss, not just the company-wide number.

Development and learning activity drops off. Someone who's still invested in their future here keeps showing up for growth opportunities. Someone who's quietly checked out stops opting in, well before they stop showing up for the day job.

One-on-ones get cancelled, shortened, or go generic. A person who trusts the relationship brings you real problems. A person who's pulling away brings you status updates and not much else.

They stop putting their hand up. For the stretch project, the cross-team initiative, the thing that used to interest them. Withdrawal from opportunity is one of the clearest tells there is.

Internal moves stop being on the table. If someone's ambitious and there's genuinely nowhere for them to go inside your organisation, they'll eventually go looking outside it. That's not disloyalty. That's just what ambition does when it runs out of runway.

None of these signals is proof on its own. A quiet month happens to everyone. What matters is the pattern over time, and whether anyone is actually watching for it.

Reading the signals before the resignation

The fix here isn't more data. Most organisations already have what they need in engagement platforms, performance systems and manager conversations. The fix is treating that data as an early warning system rather than a lagging report you glance at once a quarter.

A few practical habits:

• Look at individuals, not just averages. A healthy team score can hide one or two people who are genuinely at risk.
• Pair quantitative signals with a real conversation. Data tells you where to look. It doesn't replace actually asking. Stay interviews are the tool built for exactly this.
• Train managers to notice withdrawal, not just complaints. Silence is often a bigger signal than a complaint. A complaint means someone still cares enough to raise it.
• Act on what you see, visibly. If someone reports a concern and nothing changes, you've taught them that reporting it doesn't work, which accelerates the exact drift you're trying to catch.

Development and manager relationships are two of the strongest signals on this list, which is also why they're two of the most useful places to intervene. It's a big part of what Whyser Work's Transformation tools are built to support, matching development to where someone actually is rather than leaving it to chance.

What the drift actually costs

It's tempting to treat this as a "nice to fix eventually" problem. It isn't. Every person who drifts out the door takes their knowledge, their relationships and their productivity with them, and someone has to recruit, onboard and rebuild all of it from scratch. Rather than quote you a generic figure that won't match your business, put your own numbers into our turnover cost calculator and see what the drift is actually costing you. It's a sharper conversation to have with a board than an industry-wide statistic ever is.

Frequently asked questions

Is there one signal that matters more than the others?

If I had to pick one, it's a sustained drop in an individual's engagement score, especially paired with withdrawal from development or stretch opportunities. On its own it's not proof. Alongside other signals, it's usually the clearest early warning available.

How do we tell the difference between a bad week and a real signal?

Time and pattern. A single low score or a skipped one-on-one is normal life. A trend that holds for two or three months, especially in someone who was previously engaged, is worth a direct, honest conversation.

Our engagement scores look fine. Could we still be missing something?

Yes, easily. A healthy average can hide individual decline, particularly for your strongest people, who often mask disengagement well until they've already decided to go. Look at individuals as well as the aggregate.

What should a manager actually do when they spot the signal?

Have a real conversation, not a scripted check-in. Ask directly how the person's finding things and what would need to change. Then, and this is the part that's often missed, follow through visibly on whatever you agree to.

If you had to guess right now which of your best people is quietly drifting, would a name come to mind, and when did you last actually ask them how they were going?