# The Counteroffer Decision, Scored Against Why You Were Leaving

*You will score your specific counteroffer across the dimensions that predict regret and reach a stay-or-go decision you can defend to yourself and both employers.*

- Canonical URL: https://www.refolk.ai/candidates/guides/counteroffer-decision-scored
- Pillar: Offers and negotiation
- Format: Framework
- Published: 2026-08-07
- Last reviewed: 2026-08-07
- Reading time: 17 min

Your employer countered your resignation and you have days, not weeks, to decide. This guide is for a candidate holding both a new offer and a counteroffer who wants to grade their specific case rather than obey a slogan. It gives you a scored model across the dimensions that predict regret, so you can reach a stay-or-go call you can defend to yourself and to both employers.

The internet's answer to this question is "never accept a counteroffer," and it rests almost entirely on one statistic that has no floor under it. That is useless when your real case might be the exception. So I am going to replace the blanket rule with a scoring model built on three things: how big the gap is that the counteroffer closes, whether it touches your actual reason for leaving, and what it costs you in trust. You will end up with a number you can stand behind.

## Why the "never accept" rule fails you

The famous statistic behind every "never accept" listicle has no traceable dataset, so your real case cannot be an exception to a rule that was never measured. The most-repeated version claims 80 percent of people who accept a counteroffer leave within six months, or 90 percent within a year. Recruiter Ken Davies traced it and concluded that of "real, statistically robust data of a systematic study across a large population there is absolutely none." An independent write-up found the claim "runs out when you try to trace it, gets attributed to unnamed industry research and to a Wall Street Journal article nobody can produce, with no sample size, methodology, publication date or dataset."

Where a body is named, the attribution is inconsistent. Some pages credit the figures to SHRM, others to a firm called "Stanton House," others to a software company called "Eclipse." None produces the underlying data. That is the signature of source-laundering: a number passed hand to hand until it feels true.

There is one traceable adjacent figure. HBR cites CEB data that "50% of employees who accept a counteroffer leave within 12 months." That is a real number with a named source, and it is far less alarming than the folklore. Notice also that even the honest figure describes a population, not you.

**50% - Counteroffer accepters who leave within 12 months (CEB, via HBR)**

The one traceable figure behind a debate otherwise run on numbers with no dataset.

The point of this section is not to tell you counteroffers are safe. It is to strip away the false certainty so you can do the actual work: scoring your own offer.

> **Watch out:** The slogan is not evidence
>
> If someone quotes you the 80 percent figure, ask for the sample size and methodology. None exists. Treat any counteroffer statistic attributed to SHRM, the WSJ, or a named software firm with no link as unverified until you can reach the dataset.

## The numbers actually disagree, so name the source every time

Published counteroffer outcomes span from 32 percent to 93 percent leaving, and acceptance rates span 13 percent to 55 percent, which means opposite claims are both citable. This is not one truth being reported badly. It is different surveys measuring different populations, so any figure you use has to carry its source with it.

Here is what the named sources actually say about how many accepters leave, and over what window. The disagreement is the finding.

| Source | Left within stated window | Window |
|---|---|---|
| CEB (via HBR) | 50% | 12 months |
| Robert Half AU 2026 | 40.8% | 12 months |
| Robert Half 2026 (US) | 32% | 12 months |
| WSJ/Vault (legacy) | 93% | 18 months |
| Mallick/HBR | 80% | 6 months |

The spread from 32 percent to 93 percent tells you not to anchor on any single row. The 32 percent figure comes with useful context: the Robert Half 2026 guide reports that 85 percent of employers extended counteroffers in the past year, yet 32 percent of those employees still left within 12 months. That is roughly a two-in-three hold rate, which is a very different story from "they all leave."

Whether you even receive a counteroffer, and whether people accept, is just as contested.

| Metric | Figure | Source |
|---|---|---|
| Receive after resigning | 37.94% | grasslandsgroup.com |
| Receive after resigning | ~50% | alexanderdanielsoffshore.com |
| Accept (Robert Walters) | 13% | robertwalters.be |
| Accept (Momentum study) | 55% | momentumlegal.com |
| Small firms (10-49) making counteroffers | 80% | momentumlegal.com |

The acceptance figures of 13 percent and 55 percent cannot both describe the same population. When you read "most people accept" or "almost nobody accepts," check which survey it came from before you let it move you.

One structural fact does hold up. Counteroffer likelihood scales inversely with company size: 80 percent of firms with 10 to 49 employees make them versus 54.4 percent of firms with 250 or more. A single departure is more disruptive to a small team, so a small-company counteroffer leans toward continuity rather than a fresh read on your value. That is a clue to motive, which is the next dimension.

## The mechanism: stated reason versus actual reason

A counteroffer pays against your trigger, which is a competing number, not against your driver, which is why you started looking. When those two differ, a matched salary leaves the real problem untouched and you are back in the market within a year. This mismatch is the single best predictor of regret, and it is measurable.

A recruiter case study found that "compensation was the stated reason for departure in 70% of cases, but the primary driver in only 32%." That gap is the whole game. People say "money" because it is clean and defensible, but the actual driver is often growth, management, or culture. Nicole Gorton, a director at Robert Half, puts it directly: "money alone isn't enough to keep workers from leaving, as deeper factors like career progression, culture and engagement are at play."

> A counteroffer pays against the number that triggered your exit, not the reason underneath it.

This is why the order of operations matters. If you look at the raise first, it reframes your memory of why you were leaving. The number is concrete and immediate; the culture problem is diffuse and easy to talk yourself out of. So you write your leaving reasons down before you reread the offer. That single sequencing move is your defence against the pay-gap illusion.

#### Why a matched salary still fails

1. **Trigger** - A competing offer with a higher number arrives
2. **Resignation** - You cite compensation, the clean and defensible reason
3. **Counteroffer** - Employer matches the number that triggered the exit
4. **Driver untouched** - Growth, management or culture problem remains
5. **Re-exit** - You are back in the market inside a year

*The counteroffer lands on the trigger while the driver sits untouched underneath it.*

The three dimensions I score in this guide follow from the mechanism. Gap size measures whether the offer meaningfully closes what you were leaving over. Driver match measures whether it touches your actual reason rather than the stated one. Trust cost measures what accepting or reneging does to your standing. Score all three, because a big raise that misses the driver and a small raise that fixes it are both easy to misread.

## Score your counteroffer across the three dimensions

Grade the offer on gap size, driver match, and trust cost, then weight each by how much it matters to you. A defensible decision is one where the weighted total clears a threshold you set before you saw the number, not a gut reaction to the raise.

The scoring model is documented: "assign a 1-10 score for each priority, multiply by its priority weight, sum, compare to your baseline and external offer, and if the weighted total meets or exceeds your threshold, accept; otherwise decline." I use three composite dimensions to keep it tractable.

### Dimension 1: gap size, and what it proves

Gap size is how much of the distance between your current situation and your new offer the counteroffer actually closes. It proves the employer is willing to revalue you now. It lies when the raise is large in dollars but the driver was never money, so a big number reassures you about the wrong problem.

Score it 1 to 10 on how completely the written counteroffer closes the pay and level gap you were leaving over. A full match with a title bump is a 9 or 10. A partial match that still trails the new offer is a 4 to 6.

### Dimension 2: driver match, and what it proves

Driver match is whether the counteroffer touches the reason you actually started looking. It proves the fix is aimed at your real problem. It lies when a promised change to scope or reporting is verbal, because promises made under the pressure of a resignation fade once the crisis passes.

Score it 1 to 10 against your written leaving reasons. If your top driver was pay and pay is fixed in writing, that is a 9 or 10. If your top driver was your manager and the offer is only money, that is a 2 or 3, regardless of how big the money is.

### Dimension 3: trust cost, and what it proves

Trust cost is what your decision does to your standing with each employer and the wider network. It proves whether the path you are choosing is clean. Accepting can bring "perceptions of diminished loyalty that may limit access to future promotions" and being viewed as a flight risk. Reneging on the new employer risks recruiter memory: "recruiters track candidate behavior, and a history of reneging can lead to them being reluctant to work with that individual again."

Score it 1 to 10 on how clean the chosen path is. Declining the new offer before signing is low cost. Reneging on a signed offer is high cost, especially in a small field.

> **Rule:** Get it in writing before you score
>
> Do not assign a driver-match score to any promise about title, scope, or raise timing that lives only in conversation. If it is not in writing with dates, it scores as if it does not exist.

## The stay-or-go procedure, in order

Run these seven steps in sequence over your one-to-three-day window. The sequence matters: you buy time, get terms in writing, and write your reasons down all before the raise gets to reframe your judgement.

#### From resignation countered to defensible decision

1. **Buy an explicit window** - Ask your manager for one to three days and agree a specific follow-up date. Done when a date is on both calendars, because the documented failure mode is an emotional same-day yes.
2. **Get the terms in writing** - Require salary, title, scope, start conditions and any milestone dates in writing before you evaluate anything. Done when nothing about the offer lives only in a spoken promise, because verbal commitments about role and growth fade once the crisis passes.
3. **Diagnose the employer's motive** - Decide whether the counteroffer is recognition of your value or avoidance of the disruption your exit causes. Done when you can name which, since small firms counter at 80 percent largely for continuity rather than valuation.
4. **Write your leaving reasons before you reread the number** - List every driver that started your search: pay, growth, management, culture. Done when the list is written, because the compensation reason is stated in 70 percent of exits but is the actual driver in only 32 percent.
5. **Score each driver and weight it** - Give each driver a 1 to 10 score for how well the counteroffer and the new job each fix it, multiply by that driver's weight, and sum. Done when you have two weighted totals plus your no-offer baseline to compare against.
6. **Gate on how you resigned, then decide against a threshold** - Check whether your resignation was calm with real notice or an ultimatum, then compare the weighted totals to a threshold set in advance. Done when you have a defensible stay-or-go call, not a gut reaction to the raise.
7. **Communicate professionally either way** - Decline via the same channel your employer used, with genuine thanks, or if staying, address the original issues directly with your manager. Done when the relationship is left warm and the root causes are on the table, not buried.

Sources disagree on the order of the last two dimensions. The Investment Insight model scores first and decides against the total. The practitioner framing from Ardent front-loads one gating test: how you ran the resignation. "It depends on how you ran the resignation more than on the counteroffer itself - a calm resignation with real notice is usually unremarkable to come back from, an explicit ultimatum is not." I fold both in: score the three dimensions, but treat an ultimatum resignation as a gate that caps your realistic stay score no matter how good the number is.

Here is the scoring template, ready to copy.

**Counteroffer weighted score sheet**

```
Driver (weight /10)          | Counteroffer score | New job score
---------------------------- | ------------------ | -------------
Pay and level      (w = __)  | ____ x __ = ____   | ____ x __ = ____
Growth / progression (w=__)  | ____ x __ = ____   | ____ x __ = ____
Management / manager (w=__)  | ____ x __ = ____   | ____ x __ = ____
Culture / engagement (w=__)  | ____ x __ = ____   | ____ x __ = ____
Trust cost of this path (w=__)| ____ x __ = ____  | ____ x __ = ____
                             |                    |
WEIGHTED TOTAL               | ______             | ______
Baseline (stay, no offer, no raise): ______
Decision threshold (set before reading the number): ______
Gate: was the resignation calm with real notice?  Y / N
```

*Weights are yours and should sum to 10. Score the counteroffer and the new job 1-10 on each row, multiply by the weight, and total each column.*

The matrix below is the quick read once you have the two big dimensions.

#### Gap size against driver match

Horizontal axis runs from Driver missed to Driver fixed. Vertical axis runs from Small gap closed to Large gap closed.

| Quadrant | What it means |
| --- | --- |
| Small gap, driver missed | Decline; the offer changes nothing that mattered |
| Small gap, driver fixed | Weigh the new job; the fix is real but under-priced |
| Large gap, driver missed | The regret quadrant; money reassures you about the wrong problem |
| Large gap, driver fixed | Strongest case to stay if the fix is in writing |

*A big raise that misses your real driver is the classic regret quadrant.*

Once you have written your leaving reasons and scored the offer, the honest next question is whether your market is deep enough that walking away costs you little. That is what your negotiating leverage rests on. [Refolk](/candidates) builds your resume from your own history and scores how well you fit each posting, so you can see how many real openings sit behind your stay-or-go call rather than guessing.

## How this decision goes wrong

The predictable failures are a short list, and most of them are cases where a real signal lies to you. Learn what each looks like when it deceives, and you can catch it before you sign anything.

- **Obeying the "80 percent" slogan.** The false positive is treating folklore as data and declining a genuinely good counteroffer. Check: demand the sample size and methodology. None exists.
- **The pay-gap illusion.** Money closes the gap but your leaving reason was growth or management. Check: the 70-percent-stated versus 32-percent-actual split. Write your reason down before you see the number.
- **Verbal promise decay.** Title and scope promises evaporate once the crisis passes. Check: refuse to decide until every promise is in writing with dates.
- **Ultimatum residue.** A resignation delivered as a threat marks you as a leverage-only negotiator, and staying under that shadow is different from staying after a calm exit. Check: was your notice calm with real timing, or an ultimatum?
- **Flight-risk relabeling.** Accepting can quietly cost you the next promotion because you are now seen as someone who was one offer away from leaving. Check: ask whether you would honestly be first considered for advancement now.
- **Reneging blowback, mis-sized in both directions.** Assuming a blacklist exists is one error; assuming zero cost is the other. Check: how tight is your industry, and is there a signed contract? Reneging means you formally accept a signed written offer, then revoke it for a better one. That carries more weight than declining before signing.
- **Source-laundering.** A figure credited to SHRM, the WSJ, or "Eclipse" with no link. Check: if you cannot reach the dataset, treat it as unverified.

> **Note:** The trust cost is asymmetric and the network is finite
>
> Declining a new offer before you sign is low cost. Reneging on a signed offer is higher, because recruiters remember. In Refolk's index there are about 89,325 US recruiters against 348,418 US software engineers, roughly one recruiter per four, so in a specialised niche the same people recur across your whole career.

The relationships you keep warm are the ones you may need again, and the network is smaller than it feels. Depth of your talent pool is also your leverage: it tells you how replaceable you are and how quickly you could search again if the counteroffer sours.

| Segment | Profiles | Derived |
|---|---|---|
| US Software Engineer | 348,418 | baseline |
| UK Software Engineer | 42,941 | US pool is 8.1x the UK pool |
| US Recruiter | 89,325 | about 1 recruiter per 3.9 US engineers |

The one-recruiter-per-four-engineers ratio is why the trust dimension is not sentimental. It is arithmetic about a small, recurring set of people who move candidates.

## Verify before you commit either way

Before you send a yes or a no, run this checklist. It is the difference between a decision you can defend in a year and one you rationalised in an afternoon.

#### Before you tell either employer

- [ ] I asked for an explicit window and have a follow-up date agreed
- [ ] I have the counteroffer in writing, including salary, title, scope and any milestone dates
- [ ] I wrote my original leaving reasons down before I reread the raise number
- [ ] I scored the counteroffer and the new job on all three dimensions and set my threshold in advance
- [ ] I named the employer's motive: recognition or disruption avoidance
- [ ] I checked whether my resignation was calm or an ultimatum, and factored the residue in
- [ ] I confirmed no promise I am relying on lives only in conversation
- [ ] I know whether a signed contract exists on the new offer and what reneging would cost in my field
- [ ] I have a same-channel, warm message drafted for whichever way I decide

If you accept, address the original issues directly with your manager so the same frustrations do not resurface, and confirm every promise about role or growth in writing. A verbal counteroffer fades once the moment passes. If you decline, thank your employer sincerely and keep the relationship warm, and decline via the same method they used to extend the offer.

## Keep the decision current after you make it

A stay-or-go call is not final on the day you make it; it has follow-up conditions you should track. If you stayed, the model gives you a test: within a quarter, ask whether the driver you scored got fixed in reality, not just on paper. If growth was your driver and no new scope has arrived, you are in the population the honest CEB figure describes, and you should reopen your search deliberately rather than drifting.

If you walked, keep the counteroffer employer's relationship intact. You declined; you did not renege, since you never signed with them. That distinction is worth protecting because, as the recruiter ratio shows, the same people recur.

To pressure-test the flight-risk concern before you decide, look at how the pattern actually plays out for people in your role and market. Searching for real cases beats trusting a slogan.

Ask me this: `Senior backend engineers in London who left a company within 12 months of a promotion` - [run the search](https://www.refolk.ai/start?q=Senior%20backend%20engineers%20in%20London%20who%20left%20a%20company%20within%2012%20months%20of%20a%20promotion).

*Returns real profiles you can read to see how often a stay-after-resignation actually held or unravelled in your market.*

The whole point of scoring rather than obeying is that your case is specific. The gap you were closing, the driver underneath it, and the network you sit inside are yours. Grade them, set your threshold before the number can move you, and you will make a call you can explain to yourself, to the employer you are leaving, and to the one you are joining.

## Frequently asked questions

### Should I accept a counteroffer if it matches my new offer?

Match the number to your reason for leaving, not to the competing offer. If pay was genuinely the whole problem, the fix is in writing, and you would take the job at that number even with no outside offer in hand, a matched counteroffer can hold. If your real driver was growth, management, or culture, the raise closes the wrong gap and the frustration returns; a recruiter case study found compensation was the primary driver in only 32 percent of exits despite being stated in 70 percent.

### Is it true that 80 percent of people who accept a counteroffer leave within six months?

There is no robust dataset behind that figure. Recruiter Ken Davies traced it and found no statistically robust systematic study, and it gets attributed inconsistently to SHRM, a Wall Street Journal article nobody can produce, and software firms. The one traceable adjacent figure is CEB data via HBR showing 50 percent of counteroffer accepters leave within 12 months. Treat the 80 percent as folklore and score your own case instead.

### How long can I take to decide on a counteroffer after resigning?

Ask for a couple of days with an explicit follow-up date. Every procedural source leads with this because the documented failure is an emotional same-day yes. A short window also gives you time to get the terms in writing, which is the step that separates a real offer from a promise that fades once the crisis passes.

### Will I get blacklisted if I renege on the new job to stay?

A formal blacklist is contested; one finance source says they have never seen evidence one exists. The real cost is recruiter memory in a finite network. In Refolk's index there are roughly 89,325 US recruiters against 348,418 US software engineers, about one recruiter per four, so in a specialised niche the same people recur. Reneging on a signed offer carries more weight than declining before you sign.

### Why did my employer make a counteroffer at all?

Replacing you is expensive and disruptive, so a counteroffer is often a short-term fix that buys time to find your replacement on their schedule. Company size predicts it: 80 percent of firms with 10 to 49 employees make counteroffers versus 54.4 percent of firms with 250 or more, because a single departure hits a small team harder. Diagnose whether the offer is recognition or disruption avoidance before you weigh it.

---

*From the Refolk guide library. I revise these guides rather than replacing them, so the current version is always at https://www.refolk.ai/candidates/guides/counteroffer-decision-scored*