The Retention Counter-Offer, Scored to Stay, Leverage, or Decline

Your employer just countered your resignation with more money, a title, or both. This guide is for the candidate who has to decide, in the next few days, whether to stay, use the counter as leverage on a competing offer, or turn it down. It gives you a scoring model that grades the counter against the reason you resigned, so you reach a defensible call in under an hour instead of relying on the "never accept" folklore that dominates every search result.

The dominant advice online is an opinion propped up by an unsourced statistic. I am going to replace both with what credible data actually shows and a rubric you can apply to the case in front of you.

What a retention counter-offer decision actually is

A retention counter-offer decision is a case to grade, not a moral verdict handed down in advance. Your employer has offered to change your compensation, title, or working conditions to keep you after you announced you were leaving, and your job is to score that offer against the specific reason you tried to leave.

This is common. In the Achievers manager survey, roughly 67.5% of managers have extended counteroffers to employees who announced their intention to leave, and 55% of employees accepted. A 2018 Robert Half survey of 5,500 hiring managers found 58% make counteroffers to retain staff who receive outside offers. A UK survey found 57% of workers have received a counteroffer at some point. So you are not in a rare or shameful position. You are in a routine one, and it deserves a routine method.

The trap is that most guidance skips the diagnosis and jumps to a verdict. The verdict is usually "never accept," and it is usually backed by a number that does not exist. Before you can score anything, you need to know which numbers are real.

The real numbers, and where the folklore comes from

The honest answer to "how many people leave after accepting a counteroffer" is that credible sources bracket a range, and the width of that range is the point. The single figure with a named research entity behind it is CEB, later Gartner, cited in Harvard Business Review in 2016: 50% of employees who accept a counteroffer leave within 12 months. Other credible surveys land materially lower or higher depending on geography and window.

Compare the sources side by side and the "law" dissolves into a spread.

Geographies and methods differ across these rows, so do not read them as one measurement. Read them as evidence that the outcome depends on the situation, not on a universal rate. Retention after a counter varies enormously across credible sources, and that variance is the signal: the outcome tracks how you resigned and whether the root cause was money, not the counter itself.

The famous "80% leave within six months" claim is not on this table because it does not belong on any table. It has no traceable primary source. One of the most-cited origins is Eclipse, a recruitment CRM vendor, not a research body, with no methodology, sample size, or study period disclosed. Recruiter Ken Davies did the thing almost nobody bothers to do and went looking for the source, documenting years of searching and finding nothing statistically robust behind it.

There is a second real signal in the data that has nothing to do with attrition: whether you get countered at all depends heavily on your employer's size.

Small teams counter about 1.47 times as often as large ones, because a single departure lands hard on a company without a bench, so they react rather than pre-empt. If you are at a 30-person startup, expect a counter. If you are at a 5,000-person enterprise, a counter is a real signal that someone chose to spend political capital on you.

The dimensions that predict regret

Three dimensions do the work, and each one lies in a specific way if you read it carelessly. The whole model is: grade the counter on root-cause fit, on documentation, and on standalone acceptability, then check for durability red flags.

Here is what each dimension proves and what it looks like when it deceives you.

Root-cause fit: does the counter touch the reason you left

This proves whether the money is a fix or a stopgap. If pay was genuinely the whole problem, a pay counter fixes it. If pay was a proxy for workload, a manager, or stalled growth, the counter scores well on a problem you do not have and zero on the one you do.

The tell that this dimension is lying: the counter is money-only and your resignation reason was not money. If someone took the time to apply for new work, it is often no longer about the money, and the employee is probably still unhappy and likely to leave in the near future. Robert Half's Nicole Gorton frames the same point: money alone is not enough because career progression, culture, and engagement drive exits.

Documentation: is the fix on paper or in the air

This proves whether the lever is real or reversible. Under at-will employment, verbal promises of promotion or raise are weak and generally difficult to enforce without written evidence. One employment lawyer's framing: they offered you the job, but they have the right to rescind it, because you are an at-will employee and they can hire, fire, promote, and demote as they see fit.

The tell that this dimension is lying: "my manager is honest, so I trust the promotion is coming." Honesty is not the issue. Reorganizations, budget freezes, and manager departures are. The narrow legal exception is promissory estoppel, where you relied on a promise to your detriment and took a tangible loss, but that is a lawsuit, not a plan. Convert every lever into a countersigned document with an effective date or score it as air.

Standalone acceptability: would you take this with no outside offer

This proves whether you want the job or just want out of the discomfort of leaving. The cleanest test in the literature: accept when you would still take the job at that number with no outside offer in hand. If the only thing making the counter attractive is that it ends an awkward week, that is not acceptance, it is avoidance.

Score your counter, step by step

Run these seven steps in order. The whole pass takes under an hour of your own time once documentation comes back, and the ordering matters: you name why you resigned before you score, so a money fix cannot flatter its way past a non-money problem.

Negotiation writers usually put "are you prepared to walk" first. Recruiter blogs skip straight to "never accept." I front-load the root-cause diagnosis instead, because both other orderings hide the one question that determines whether the money can possibly work: what were you actually leaving.

Durability signals: reading whether the counter holds

A counter can score well today and still be hollow, so before you commit, test three durability signals. Generosity and durability are unrelated: Robert Half found 85% of employers extended counteroffers, yet 32% of those employees still left within 12 months, because money addresses the priced problem and leaves the unpriced one untouched.

• Timing. Did the counter appear only after you gave notice? A raise that required a resignation to trigger tells you the market, not your work, sets your pay there. Expect to have to threaten to leave again next time.
• "Top of range" language. If the matched raise comes with "you are already at the top of the band," you may not have gotten a raise so much as next year's raise, early. Where that is how it works, the follow-on increase is small and pre-explained. Check what next cycle's raise looks like before you count this as real growth.
• Root-cause silence. If your resignation reason was your manager, workload, or growth, and the counter is silent on all three, the durability is whatever your patience is. Counteroffers frequently fail to resolve underlying issues like career stagnation or poor management.

A generous counter and a durable counter are different animals; only one of them survives contact with next Tuesday.

The leverage play, and how to run it without reneging

Leverage means asking your employer to improve their terms because you have a credible outside offer, and it works only when the alternative is genuinely credible. The governing rule from negotiation research is blunt: do not use an external offer as leverage unless you are genuinely prepared to accept it. If the other party believes you will not actually walk away, much of that power disappears.

This is where the market matters, and it is where your position is stronger than the folklore admits.

In Refolk's index of professional profiles, there are roughly 349,715 US software engineers against 13,908 recruiters, about 25 engineers per recruiter. Read that as depth of outside options: a resigning engineer sits in a pool deep enough that "prepared to walk" is believable. Credible alternatives are what make the negotiation rule work, and they are exactly what a healthy market supplies.

The etiquette has three hard edges:

• Make it a request, not a demand. An ultimatum can make your boss feel cornered and tell you to move on. Ask; do not threaten.
• Do not lie about a job offer. Bluffing can damage your reputation, and they may call your bluff and show you the door.
• Do not disclose the exact competing number. If your offer is lower than the highest your employer would have paid, they now know they never had to go that high, and you undercut yourself.

If you want to pressure-test how deep your outside pool actually is before you lean on it, a search across public profiles will show you the real mobility in your market.

I want to be straight with you: I have a written offer elsewhere, and I would rather stay if we can close the gap on the things that made me look. The two that matter most to me are [progression / scope] and [compensation]. Could we look at what is possible here? I am asking, not issuing an ultimatum - I would like to make this work if we can.

How this goes wrong

The scoring model fails in predictable ways, and each failure has a false positive that feels like good judgement in the moment. Learn the tell for each.

• Citing the 80% stat as fact. It sounds researched, which is the trap. Check for sample size, methodology, and publication date. There is none: it is folklore repeated long enough to sound like evidence.
• Scoring the counter without naming why you resigned. A money fix scores well against a non-money problem. If pay was not the whole cause, the raise is a stopgap no matter how large.
• Treating a verbal promotion as secured. "My manager is honest" is not the test. The test is whether it is on countersigned paper with an effective date. If not, it is reversible.
• Bluffing with an offer you will not take. If your honest answer is that you would immediately turn down the other job and stay, you never had much leverage in the first place.
• Disclosing the exact competing number. "Transparency builds trust" feels virtuous and costs you money. Revealing a figure below their ceiling tells them they do not have to reach it.
• Reneging on a signed agreement without reading it. At-will does not always mean free exit. Check for notice periods or penalty clauses first.
• Reading a matched raise as a real raise. A "top of range" note signals you pulled future money forward. Check next cycle's increase before you celebrate.
• Assuming the counter is durable because it was generous. 85% of employers extended counteroffers yet 32% of those employees still left within 12 months. Generosity is not durability.

What to check before you call it done

Whichever way you decide, verify the case is actually closed rather than half-decided. This is the list I would run before telling anyone your answer.

Keeping the call defensible over time

The numbers in this guide are ranges, not constants, so the way to keep your decision current is to re-run the diagnosis rather than to trust a remembered figure. When your situation changes - a new manager, a reorganization that erases the promotion, a comp cycle that reveals the "top of range" note was real - your score changes, and the honest move is to score it again.

Two mechanisms are worth re-checking locally rather than taking on faith. First, documentation: a lever is only as durable as its paper, so if the countersigned memo never arrived, the counter never fully existed and your standalone-acceptability answer should default to no. Second, the market depth behind your leverage: outside options are what make "prepared to walk" credible, and they shift by role, place, and stage. Refolk tailors your resume to each posting and scores how well you actually fit, which is the fastest way to convert "I could probably leave" into a concrete alternative you can hold up against the counter. When you are weighing a counter, the most useful thing you can do is make the outside option real enough to test the three-yes rule honestly - and if you want that on tap, Refolk is built to write and tailor the applications that keep a credible alternative live.

The counter is a case. Grade it against why you left, put every lever on paper, and let the score, not the folklore, make the call.