I was the HR manager, and the leadership team had a problem. Key people were leaving, the kind you don’t easily replace, and nobody could say why. So they came to me with a plan. They wanted exit interviews, and they wanted engagement surveys. Let’s find out what’s going on.
I could give them both. But I told them neither one was going to answer the question they were actually asking.
Here’s the thing about an exit interview. By the time someone is sitting across from me telling me why they’re leaving, the decision is already made. They’re gone. I’m collecting the reason after I’ve lost the person, not in time to keep them. It’s useful, but it’s an autopsy. Nobody’s life got saved in an autopsy.
And engagement surveys? Let’s be honest about what those usually are. They’re satisfaction surveys. People anonymously tell you how they feel about their manager, their benefits, their workload, the technology, the leave policy. That gives you a story about the group, and that has real value. But it comes with two big caveats. One, it’s anonymous, so you can’t act on it for any single person. Two, people often aren’t honest in writing, and they definitely won’t put the specific, risky stuff down on a form. So you end up with a mood ring for the whole company and nothing you can use for the one employee you’re trying not to lose.
So I said, why don’t we look at what we already have? The trends are sitting right there in the system.
For your key people, pull the history:
- When were they last promoted, or even considered for one?
- What’s the trend on their pay increases and bonuses?
- What do their performance reviews look like next to that pay and promotion history?
- Have their responsibilities grown without the title or the grade catching up?
- When were they last recognized for anything?
- Is their overtime climbing? Is their unused PTO piling up?
- Have they quietly stopped volunteering for projects, training, or stretch work they used to raise their hand for?
So we looked. And there it was. Some of our people had been taking on more and more work, but hadn’t moved. Not just that they weren’t made managers, they hadn’t moved a single step inside their own compensation grade. Level one, still level one. More responsibility, no promotion, no new title, just a bigger pile of work.
And you could see it in how they showed up. Their own self-reviews were getting shorter. Less enthusiastic, less detailed, year over year. The person who used to write three paragraphs about what they’d accomplished was now writing two lines. That’s not laziness. That’s someone checking out.
None of that was a secret. It was all in the system. We just hadn’t been looking at it as a pattern, and in some cases we hadn’t been recording it at all. That’s the real catch. If your system isn’t tracking promotions, grade movement, recognition, and the shape of your review data over time, you won’t have this. But if it is there, it will tell you something a survey never will. It will tell you who’s about to be out the door, while there’s still time to do something about it.
Keep your surveys. Do the exit interviews. They have their place. But stop treating them as your early warning system, because they aren’t one. The early warning system is the data you already own. You just have to look at it before the resignation lands, not after.