"I've got something else"
She's been here two years, she trains the new starters, and she has just told the store manager she's leaving in four weeks. She's apologetic about the timing. She has a job at the place across the car park.
He has about twenty minutes and three instincts, and two of them are wrong.
The first is to ask what they're paying her. The second is to say he'll see what he can do. The third — the one worth having — is to find out what actually happened, because whatever it is has almost certainly happened to three other people who haven't resigned yet.
The conversation is for information, not reversal
Most retail employee retention conversation advice is really counter-offer advice: find the number, match it, keep the person. It performs badly, and in an hourly retail operation it performs worse than average.
Four things make this specific.
The lever usually isn't yours. Pay bands are set centrally. A store manager who implies he can fix the rate is writing a cheque on someone else's account, and when it doesn't clear he has lost the relationship and the four weeks.
It's the end of a process, not the start. By the time someone hands notice in they've applied, interviewed, been offered, and told someone at home. The decision was made weeks ago; the resignation is the paperwork.
You'll keep managing them either way. Unlike a customer save call, this counterparty is on your rota tomorrow. Handle it badly and you lose the last four weeks as well as the person — the handover, the training, the goodwill with everyone watching how you took it.
And the information is the asset. She knows exactly why she's going. That is the most accurate operational feedback available to a store manager, it is free, and it has a shelf life of about twenty minutes.
Why the counter-offer is the wrong move
It is the most common response and it fails in three distinct ways.
It usually doesn't hold — a large share of people who accept one leave within the year anyway, because the reason was rarely only money and the underlying thing is still there.
It creates a pay-equity problem immediately. The colleague doing the same job on the original rate finds out, because they always find out, and the operation has just taught its team that the route to a rise is a resignation letter.
And it answers the wrong question. If she's leaving because the rota goes up on Sunday for a Monday start, an extra forty pence a week doesn't touch it — but the rota is something a store manager can genuinely change, and nobody will mention it if the conversation is about money.
What to ask
1. Accept it first, properly. "Right — I'm sorry to hear it, and congratulations." Anything that sounds like negotiation in the first minute closes the useful part down.
2. Ask when she decided. "When did you start looking?" The answer is the diagnostic. Three weeks ago is a specific event. Since February is a condition.
3. Ask what would have had to be different. Not "why are you leaving" — which gets the safe answer — but the counterfactual, which is harder to deflect and usually produces something concrete.
4. Separate the push from the pull. What's better over there, and what was wrong here, are different lists. The pull is information about the market; the push is information about you, and it's the one that matters.
5. Say plainly what you can and can't do. "I can't move the rate — that's not mine. I could have changed the shift pattern, and I didn't know it was the problem." Honest, and it treats her as an adult in a way a vague "let me see what I can do" doesn't.
6. Don't guilt-trip, and don't go cold. "After everything we've done for you" is the sentence the whole team hears about by Thursday. So is a manager who stops speaking to someone on their notice.
7. Leave the door open, specifically. Retail boomerangs are common and cheap to rehire. "If it's not what you're hoping, call me" costs nothing and is remembered.
8. Then do the part that's actually the point. Take what she told you and have the same conversation — before anyone resigns — with the two or three people you'd least like to lose.
The version that works is the one held in March
Everything above salvages information from a decision already made. The intervention that changes an outcome is the same set of questions asked of someone who hasn't resigned: what would make you leave, what's the most annoying part of this job, what would you change about the rota.
It takes fifteen minutes, twice a year, per person you'd rather keep. Almost no retail operation does it, and it's the single cheapest retention mechanism available to a store — it needs no budget approval, which is precisely why nobody has built a process around it.
Four ways it goes wrong
The counter-offerer, who goes straight to money — usually without authority, often unsuccessfully, and always visibly to the rest of the team.
The guilt-tripper, who makes it personal. Converts a good leaver into a bad reference in both directions.
The offended manager, who takes it as a betrayal and goes cold for four weeks, losing the handover and teaching everyone what resigning here costs.
The note-taker, who asks all the right questions, writes them down, and changes nothing — which is worse than not asking, because she'll tell the others he asked.
Why this isn't trained
Exit interviews are owned by HR and happen too late. A form, a fortnight later, with someone she's never met. The answers are polite and useless, and the manager who could have acted never sees them.
Retail turnover is normalised. At high baseline attrition, individual resignations stop being treated as events, so nobody diagnoses them. The rate is reported monthly and explained as "the sector."
Managers are told about resignations administratively. It arrives as a rota problem and a requisition, not as a conversation with a window on it.
And peer role play gives an honest answer immediately. A colleague playing a leaver explains their reasons clearly in the first minute. A real one says "it's just a bit closer to home" and means something else entirely — and getting past that is the whole skill.
What retention conversation training can rehearse
A simulation can hold a leaver who gives the safe answer first and the real one only to a manager who asks well — which is the actual difficulty and the thing a form can never capture. Foretell AI provides the counterparty configuration, transcripts and rubric-based scoring; pay authority, retention policy and the exit process stay with the retailer.
Four to build:
- The polite deflector, whose stated reason is commute or hours and whose real reason is a person or a rota.
- The one who wants to be talked out of it, testing whether the manager can hear that without immediately offering money.
- The one with a specific, fixable grievance that was raised once, months ago, and went nowhere.
- The stay interview, with someone who hasn’t resigned — the version that matters and the one managers find most awkward.
Designing the module
Pass one — the opening. Score whether the resignation was accepted before anything was offered, and whether money was raised in the first two minutes.
Pass two — the diagnosis. Score whether the manager asked when the decision was made, asked the counterfactual, and separated push from pull.
Pass three — the stay interview. Same questions, no resignation, someone who hasn't thought about leaving in words yet.
Rubric on observable behavior: Was the resignation acknowledged before negotiation? Was a counter-offer made, and with what authority? Was the timing question asked? Was a specific, fixable cause identified? Was anything promised that the manager can't deliver? Was the door left open?
Whether a specific cause was identified is the measure worth tracking, because it's the only output of this conversation that can change anything for anyone else.
The operator case
Replacement cost is the comparison nobody puts in front of the manager. Recruiting, onboarding and bringing an associate to full productivity costs a multiple of the adjustment that would have kept them, and the manager making the call almost never sees that arithmetic.
Counter-offers are a pay-equity exposure. An off-band rise granted under pressure is visible within days and teaches a route that other people will take.
Regretted attrition isn't measured on hourly roles. Most retailers track turnover as a single rate. Distinguishing the leavers you'd have paid to keep from the ones you wouldn't is a small reporting change and it redirects the entire conversation.
And the stay interview is a free intervention with no owner. Fifteen minutes, twice a year, for the people you'd least like to lose. It requires no budget, which is also why nobody has been made accountable for it.
For retail management programmes, this is a clean lesson in the difference between a decision and its announcement: the resignation is not the moment the employee decided, and treating it as one guarantees you intervene too late.
Frequently asked questions
Should a manager make a counter-offer when someone resigns? Usually not. Acceptance rates are poor over twelve months, the reason is rarely only money, and an off-band rise creates a pay-equity problem and teaches the team that resigning is the route to one.
What should you ask when an employee resigns? When they started looking, what would have had to be different, and what's better about the new role versus what was wrong with this one. Those three separate a fixable cause from a market move.
What is a stay interview? The same conversation held with someone who hasn't resigned — what would make you leave, what's most frustrating, what would you change. Fifteen minutes, twice a year, and it's the version that actually changes outcomes.
How do you stop good retail staff leaving? Mostly by finding out what's wrong before they've been offered another job. Once notice is handed in, the decision is weeks old and the conversation is worth more as information than as a rescue.
The short version
By the time she tells him, she's already told her partner, her new employer and probably two colleagues. What's left isn't a negotiation — it's the most honest twenty minutes of operational feedback he'll get all year, and the counter-offer is the fastest way to waste it.
Accept it, ask when she decided, ask what would have had to be different, say honestly what you can and can't change, leave the door open.
Then go and ask the same questions of the three people who haven't resigned. That's the conversation that was always the point.
Foretell AI lets retailers build conversational simulations — including resignation, stay interview, and retention conversations like the one above — with configurable counterparties, transcripts, recordings, and rubric-based evaluation. If your exit interviews arrive at HR two weeks after the manager could have acted, we're happy to walk through how other operators have structured it.