Why good employees leave: the conversation nobody had

Good employees most often leave because they do not feel valued and cannot see a clear future, not because of pay. Gallup found that 51% had no conversation with any leader about their satisfaction or their future in the three months before they left, and 52% say the departure could have been prevented. The intervention is that conversation.

The worst part of a resignation is not the resignation. It is the surprise. Someone walks into your office, mentions an offer, and you realise you saw none of it coming. You had dozens of conversations with that person over the past six months. All of them about projects, deadlines, clients. None of them about the person.

Then the reflex kicks in: the counteroffer, the raise, the promise of a better project. Sometimes it buys a few months. Usually it buys nothing, because you are answering a question your employee asked themselves a long time ago and has already answered.

Why good employees leave is a question HR rarely gets to diagnose properly. Not for lack of data, but because we look at the data at the wrong moment: after the exit, when the only thing left to learn is who is next.

TL;DR

• The real reasons are relational, not financial: in a McKinsey survey of 5,774 employees, the top three reasons given for quitting were not feeling valued by the organisation (54%), not feeling valued by their manager (52%), and not feeling a sense of belonging at work (51%), while employers cited compensation and work-life balance.
• The window closes before the resignation. By the time it arrives, everything after is negotiation, not retention.
• Silence is the signal, not conflict: 51% of voluntarily exiting employees say that in the three months before they left, neither their manager nor any other leader spoke with them about their job satisfaction or their future with the organisation (Gallup, 2019).
• More than half of departures were preventable: 52% of voluntarily exiting employees say their manager or organisation could have done something to stop them from leaving (Gallup, 2019).
• The retention conversation: four questions, twenty minutes, once a quarter, run by the direct manager with every person on the team. It is not a performance review, and it does not work once someone has already announced they are leaving.
• The exit interview is not a retention tool, it is an organisational learning tool. If the only time you ask someone what is on their mind is the day they leave, you have built a system that always finds out too late.

Why do good employees leave?

Because they do not feel valued and cannot see a clear future where they are, not because someone offered 15% more elsewhere. Money is the reason they give on the way out, because it is the least confrontational answer available. Research consistently shows a wide gap between what employees cite and what employers assume.

The numbers show exactly this gap. Surveying 5,774 employees and 250 talent managers across five countries, McKinsey researchers found that the top three reasons employees gave for quitting were that they did not feel valued by their organisations (54%), did not feel valued by their managers (52%), or did not feel a sense of belonging at work (51%). When the same researchers asked employers why their people had quit, employers cited compensation, work-life balance, and poor physical and emotional health (De Smet, Dowling, Mugayar-Baldocchi and Schaninger, "'Great Attrition' or 'Great Attraction'? The choice is yours", McKinsey Quarterly, 2021).

Put simply: employees talk about relationships, employers respond with transactions. The survey covered Australia, Canada, Singapore, the UK and the US, so it does not transfer automatically to every market. But the pattern we hear in diagnostic interviews with Romanian and Hungarian teams looks familiar: people rarely start with salary, even when that is where they end up.

That reframes the problem entirely. If people left for money, retention would be a budgeting problem. Because they leave for relational reasons, retention becomes a leadership problem, which means it is built or lost in weekly conversations rather than in the annual compensation review. It is the same logic we set out when writing about what actually keeps people committed to an organisation.

What happens before a resignation?

Not one event but three stages, at least as a working model: quiet withdrawal, the decision, and the resignation. Organisations almost always intervene at the third, when the window has closed. Gallup's finding marks that window: in the three months before people left, 51% had no conversation with any leader about their future.

Stage 1: quiet withdrawal. The person still performs, but has stopped proposing. They no longer push back in meetings, no longer raise problems, no longer ask what comes next. From the outside it looks like a calm team. That is what makes this stage dangerous: silence looks like agreement.

Stage 2: the decision. It happens internally, usually without drama. The person answers one question, "is there anything left for me here?", and the answer is no. From that point on, everything else is logistics: CV, conversations, interviews.

Stage 3: the resignation. This is when the organisation finds out. This is where the counteroffer and the surprise show up. Both arrive after the window has closed.

The retention window is stage 1 and the beginning of stage 2. It is the only interval where a conversation still changes the outcome, because the question is still open. Once someone has answered it, a raise does not change the answer. It postpones the exit by a few months and adds resentment: "I had to resign to be seen."

For example, a strong specialist who asks three times, in three different meetings, to work on a new area and hears "let's talk after this project" each time will not visibly complain. They will simply stop asking. The moment they stopped asking is the moment an alarm should have gone off. It is usually the quietest moment in the whole sequence.

Why does the exit interview always arrive too late?

Because it happens after the decision has already been made. The exit interview is not a retention tool but an organisational learning tool: useful for the next person, not the one walking out. Confusing the two is why many companies hold files full of exit interviews and post the same turnover rate year after year.

Gallup's data shows how large the missed window is. Studying employees who left voluntarily, researchers Shane McFeely and Ben Wigert found that 52% of them say their manager or organisation could have done something to prevent them from leaving. And 51% say that in the three months before they left, neither their manager nor any other leader spoke with them about their job satisfaction or their future with the organisation (McFeely and Wigert, "This Fixable Problem Costs U.S. Businesses $1 Trillion", Gallup, 2019).

Read those two figures together, because together they say something the individual numbers do not. More than half of the departures were reversible. And over the same period, more than half of those people were asked nothing at all. This is not a story about managers who tried and failed. It is a story about managers who never got to try, because nobody asked them to and nothing in their calendar required it.

The cost is concrete. The same Gallup analysis estimates that replacing an individual employee costs between one-half and two times that employee's annual salary, an estimate the authors describe as conservative. In other words, you are losing far more in hard money than the conversation nobody had would have cost.

What does a retention conversation look like?

A short structured discussion, held by the direct manager, once a quarter, with every person on the team. Twenty minutes, four questions, no project agenda and no appraisal form. You hold it with people who have no intention of leaving, not with people who have already announced.

Four questions, in this order:

1. Which part of your work gives you energy right now, and which part drains it? This opens on content rather than emotion. People answer honestly about the work before they answer honestly about themselves.
2. What do you want to be doing in two years, and how plausible does it feel that it happens here? This question tests stage 2 directly. A vague or polite answer is itself an answer.
3. What have you asked for recently and not received? Not "do you have any concerns". The concrete phrasing surfaces abandoned requests, and abandoned requests are the single best predictor on this list.
4. What would need to change for you to be clearly more satisfied six months from now? Close on an action, not on a mood.

Three rules separate a retention conversation from a form. First: the direct manager runs it, not HR. HR can design the process, but the relationship is what produces the honest answer. Second: you do it while things are going well, not after someone has announced they are leaving, otherwise it becomes a negotiation. Third: if three consecutive conversations produce no visible change, people stop answering honestly and the tool becomes theatre.

The retention conversation is not an exotic capability. It is a dialogue capability that can be trained, like feedback or delegation, and it sits exactly in the territory we build in Manager OS. Good managers do not improvise empathy. They work from a set of questions and a rhythm.

What you can do on Monday

Do not launch a retention programme. Start with a twenty-minute exercise, alone, on a sheet of paper.

Write down the last three good people who left your company. For each one, answer a single question: in the three months before the resignation, who spoke with them about their future here, and what did they say? If you cannot answer that for at least two of them, you already have your diagnosis. It is not a compensation problem. It is a problem of conversations that never happened.

Then do the same in reverse: write down the good people you still have, and mark the ones nobody has spoken with about their future in the last three months. That is your risk list, and unlike the first one, you can still do something about it.

Retention is not bought at the end. It is built in the interval where someone is still asking the question, not after they have answered it. A Capability Snapshot shows you where that risk sits before you read it in resignation letters.

So the question we will leave you with is not how big the last raise was that you approved to keep someone. It is this: when did you last ask someone who has no intention of leaving what would make them stay?

FAQ

Why do good employees leave a company?

Most often because they do not feel valued and cannot see a clear future in the organisation, not because of salary. In a McKinsey survey of 5,774 employees, the top three reasons given were not feeling valued by the organisation (54%), not feeling valued by the manager (52%), and lack of belonging (51%), while employers cited compensation and work-life balance. Salary is the stated reason at exit, rarely the underlying cause.

What happens in the three months before someone resigns?

Usually nothing the organisation notices. Gallup data shows that 51% of people who left voluntarily had no conversation about their job satisfaction or their future in the three months before leaving, and 52% say their manager or organisation could have done something to prevent it. The resignation is when the organisation finds out.

What is a stay interview and how is it different from an exit interview?

A stay interview, or retention conversation, is a short structured discussion held by a direct manager with an employee who has no intention of leaving, about what motivates them and what would keep them. An exit interview happens after the decision has been made, so it can help the next employee, not the one walking out.

Do counteroffers work to retain a good employee?

Rarely over the long term. A counteroffer answers a decision made for relational reasons with a financial argument, and in many cases it only postpones the exit. Worse, it signals to the team that resigning is the channel through which things get granted, which creates exactly the behaviour you do not want.

What does losing an employee actually cost?

Gallup's 2019 analysis estimates the cost of replacing an individual employee at between one-half and two times that person's annual salary, and the authors note this is a conservative estimate that excludes lost knowledge, the effect on team morale, and customer relationships.

Sources

• Gallup. The Fixable Problem That Costs Businesses $1 Trillion. Voluntary turnover and the manager's part in it.
• McKinsey & Company (2021). Great Attrition or Great Attraction? The Choice Is Yours. The gap between why employers think people leave and why they say they left.
• Azimut internal program data: 600+ programs delivered, 20,000+ participants, 9.27/10 average trainer score (since 2008).

By Szilárd Kacsó