Should you counter-offer a resigning software engineer? Only if the reason they're leaving is something you can fix, and you should decide within 48 hours. If the reason is money you could have paid last year, matching may work. If it's the manager, the product direction or a ceiling on scope, a bigger salary mostly buys a few months. That's our judgment, and the evidence behind it is thinner than the internet suggests. This guide is written for the CTO or VP Engineering holding the decision, and it sits with our other engineering management guides.
Does a counter-offer actually keep a senior engineer?
Less often than the confident claims say. Most of the evidence is about employees in general, not engineers, so treat it as direction rather than a forecast.
The one figure we can attribute is a CEB number quoted by Harvard Business Review: 50% of employees who accept a counter-offer leave within 12 months. We read it in an HBR podcast reading list, not in CEB's original article, so we're citing HBR's quotation of it. HBR's page also carries Rebecca Knight's blunt line, "If someone has made the decision to quit, they're unhappy," and an argument from Kay and Cullen that accepting a counter-offer will adversely affect one's career.
You'll also see "80% of people who accept a counter-offer leave within six months." Don't repeat it. Momentum Legal says the figure often stems from anecdotal industry observations rather than systematic research, and we found no study behind it.
The more useful number for you is from Gallup. 52% of voluntarily exiting employees say their manager or organisation could have done something to prevent it, and over half of exiting employees (51%) say neither their manager nor a leader spoke with them about job satisfaction or their future in the three months before they left. Many resignations are fixable, and many weren't discussed.
Why did the offer arrive now?
Before you name a number, find out what changed. In our judgment, people rarely resign over a single event. They resign when a recruiter's offer lands on top of something that already bothered them.
Sort the reason into pay, scope, manager or growth. Recruiting From Scratch suggests asking "What changed?" and "Is the reason you were looking still true?" We'd use both as the opening of your conversation. Be aware that a counter-offer shows you could have paid more all along and waited until they threatened to leave. In our judgment, the engineer will notice that too.
Ask the questions once and listen. If the answer is "they're offering a lot more," you're hearing about pay. If it's "I haven't shipped anything I care about in a year," more pay won't change it.
When is matching reasonable, and when not?
Our view, by reason for leaving:
| Reason for leaving | Counter? | Why |
|---|---|---|
| Pay is below market and you can fix it | Usually yes | The cause is removable, and you'd want to fix it for the team anyway |
| Scope, title or ownership | Yes, if you can change the role | Cash doesn't address it; a real change in responsibility does |
| Manager or product direction | Rarely | The cause stays after the raise, so plan the exit |
| They're a single point of failure | Maybe, as a bridge | Buy time for a handover, not a permanent fix |
Two costs sit outside the table. The first is precedent. In our experience, engineers talk, and a raise that only appears after a resignation can teach the rest of the team how to get one. A DaedTech reader comment puts the manager's side well: "Too many managers are afraid to open a dialog about an employee's job satisfaction." If you match, fix the pay band for the role so the answer isn't "resign to get paid."
The second is fairness to people who stayed without a competing offer. If the counter puts one senior engineer well above peers at the same level, in our judgment you should expect that conversation to arrive.
On size, we looked for a benchmark for questions like "is a 20% counter-offer too much?" and found no sourced figure. Don't anchor on a percentage. Anchor on the market rate for the role and the replacement cost below.
What can you offer besides cash?
Scope, a title that matches the work they already do, and ownership of something they've asked for. These cost little and answer the scope and growth reasons directly. We cover the longer list in how to retain senior engineers without a promotion budget, so we won't repeat it here.
What does it cost to say no?
Gallup puts the cost of replacing an individual employee at one-half to two times annual salary. That's a range for employees generally, not engineers, and it moves with seniority and how specialised the role is. For a senior engineer who owns systems nobody else understands, assume you're toward the upper end (judgment).
The comparison for the decision is simple: the extra cost of the counter against that range, plus the months of reduced output while a replacement ramps up. A raise that costs a fraction of that range is cheap if it's likely to hold. If you expect them to leave within a year anyway, as the CEB figure suggests half of acceptors do, you've paid for a delay.
If you counter: write it down
Put the terms in writing: the new pay, scope or title, the date they start, and a 90-day check-in where you ask whether the original reason has gone away. Without the check-in, you won't find out until the next resignation.
Then plan the backfill anyway. Don't tell the engineer, and don't treat it as distrust. It's the same risk you carry for every key person. Keep the access register and the knowledge inventory current, so a second resignation in six months starts from a prepared position.
If they go: the first 48 hours
Decline cleanly, and don't turn the last weeks into an argument.
- Secure the accounts that only they control, and list what only they know.
- Agree the last day and what they'll hand over before it.
- Decide who covers the work, and when that person starts.
- Tell the team yourself, before they hear it elsewhere.
What this means for your retention decision
Decide on the reason, not the number. If you can name a cause you're able to remove, such as pay under the market or a role without scope, the counter is a fix and worth budgeting. If the cause is the manager or the product direction, a raise defers the exit, and the money is better spent on cover.
Budget the counter against Gallup's one-half to two times salary range, and write the 90-day check-in into the offer. Whichever way you go, keep the access and knowledge inventory running: on the CEB figure, half of acceptors are gone within a year.
If you've decided to let them go, the next step is the handover and cover plan for a key engineer who resigns, which turns the first 48 hours into a dated schedule.
FAQ
Should you counter a resigning engineer?
Counter if the reason for leaving is one you can fix and you'd be happy with the new terms for the whole team. Don't counter to avoid the work of a search. The reason matters more than the amount, and the table above sorts reasons by whether a counter is likely to hold.
Is a 20% counter-offer too much?
We found no sourced benchmark for counter-offer size, so we won't give you a safe percentage. Compare the cost of the raise with Gallup's replacement range of one-half to two times salary, and check it against the market rate for the role and the pay of the engineer's peers.
Why do so many software engineers quit?
We have no engineer-specific data. For employees generally, Gallup finds 52% of voluntarily exiting employees say their manager or organisation could have done something to prevent it, and over half of exiting employees (51%) say nobody talked to them about their future in the previous three months. That points at conversations you can have before a resignation.
Why do people say never accept a counter-offer?
Because the reasons for leaving usually persist after a raise. HBR quotes a CEB figure that 50% of employees who accept a counter-offer leave within 12 months. The better-known 80% in six months figure often stems from anecdotal industry observations, according to Momentum Legal, and we found no study behind it.
Sources
- Harvard Business Review: Should You Accept Your Boss's Counteroffer To Stay At The Organization? (podcast page)
- Gallup: This fixable problem costs U.S. businesses $1 trillion
- Momentum Legal: Statistics Reveal a Surprising Truth about Counteroffers
- Recruiting From Scratch: How to close more engineering offers at a startup
- DaedTech: Resignation counter-offer danger