Offer acceptance rate: the formula, what to count, and what a falling rate tells you
On this page
Offer acceptance rate is the share of offers that candidates accept: offers accepted ÷ offers that reached a final answer × 100. It is one of the simplest recruiting metrics to calculate and one of the easiest to get wrong, because the answer changes depending on how you count pending offers, revised offers, offers the employer withdrew, and candidates who accept and then back out.
This page sets out the counting rules, works through an invented quarter, shows how to read decline reasons, and gives a reporting format. The funnel stages that lead up to the offer are covered in pipeline conversion rates by stage.
The formula and the counting rules
Offer acceptance rate = offers accepted ÷ (offers accepted + offers declined) × 100
Counted in the period in which the candidate gave a final answer.Writing the denominator as accepted plus declined, rather than "offers extended", forces the decisions that keep the number honest:
| Situation | Rule | Why |
|---|---|---|
| Offer still pending at period end | Leave out; count it in the period it resolves | It has no outcome yet |
| Offer rescinded by the employer (failed check, role cancelled) | Leave out; count separately | The candidate did not decide |
| Offer revised after negotiation | One offer per candidate per requisition | Otherwise a negotiation counts as a decline plus an acceptance |
| Verbal offer, then written offer | Count the one you always make first; usually the verbal | That is where the candidate actually decides |
| Accepted, then withdrew before starting | Accepted in this metric; also counted as a renege | Separates offer problems from notice-period problems |
| Internal transfer or promotion | Report separately or exclude | Internal offers are accepted at very different rates |
Worked example: one quarter
Invented data: a company's Q3 offers across all roles
Offers extended in Q3: 48.
Still pending on September 30: 3. Rescinded by the employer: 2 (one failed a licence check, one role was cancelled).
Resolved by the candidate: 48 − 3 − 2 = 43. Of those, 36 accepted and 7 declined.
Offer acceptance rate: 36 ÷ 43 = 83.7%.
If the pending and rescinded offers had been left in the denominator, the rate would read 36 ÷ 48 = 75.0%, almost nine points lower for reasons that have nothing to do with candidates.
Of the 36 who accepted, 2 withdrew before their start date. Renege rate: 2 ÷ 36 = 5.6%. Offer-to-start rate: 34 ÷ 43 = 79.1%.
The three pending offers join the Q4 figures once they resolve. That means Q3's number can be reported on October 1 and will not change afterwards, which is the property you want from a metric people compare quarter to quarter.
Segment before you conclude anything
A company-wide rate hides the roles where offers are actually failing. Split it by role family, level and, if you can, by hiring manager.
| Role family (invented) | Resolved offers | Accepted | Declined | Acceptance rate |
|---|---|---|---|---|
| Customer support | 18 | 17 | 1 | 94.4% |
| Sales | 10 | 8 | 2 | 80.0% |
| Engineering | 11 | 7 | 4 | 63.6% |
| Finance | 4 | 4 | 0 | 100.0% |
| Total | 43 | 36 | 7 | 83.7% |
The 83.7% overall rate looks healthy, but four of the seven declines came from engineering, where the rate is 63.6%. That is the conversation to have. Finance's 100% is four offers; one decline next quarter would make it 75%. Show the counts next to every rate so no one reads a small group as a trend.
Reading the decline reasons
The rate tells you whether there is a problem. The reasons tell you which one. Record a reason for every decline, in fixed categories, on the day the candidate says no, while the recruiter still remembers the call.
| Decline reason (Q3, invented) | Count | Usually points to |
|---|---|---|
| Compensation below expectation | 3 | Expectations not confirmed early, or a pay band out of line with the market |
| Accepted a counteroffer | 2 | Motivation to leave not tested before the offer |
| Accepted another offer | 1 | Process too slow, or competing processes not tracked |
| Remote or location terms | 1 | Work arrangement not settled at the screen |
Almost every category traces back to something that could have been learned before the offer. Compensation declines usually mean expectations were asked once, at the screen, and never rechecked; the salary expectation questions page covers how to ask and when to ask again. Counteroffer declines point to a missing conversation about why the candidate wants to leave, which the counteroffer conversation script handles. Losses to competing offers are often a time-to-hire problem; see time to hire vs time to fill.
Avoid a catch-all "personal reasons" category. It fills up quickly and tells you nothing. If a candidate gives no reason, record "no reason given" so the size of the unknown is visible.
What raises the rate, and what only raises the number
The durable way to improve offer acceptance is to find out before the offer whether the candidate will say yes. A pre-close covers pay, start date, notice period, competing processes and anything that would stop them accepting; how to pre-close candidates has the questions. The offer call itself then confirms rather than persuades; see the offer call script for recruiters.
Some changes make the rate go up without improving anything:
- Only extending offers that are certain. If recruiters stop making offers to strong candidates who might decline, the rate rises and the team hires fewer strong people.
- Moving the counting point. Switching from verbal to written offers makes the rate jump, because most declines happen at the verbal stage.
- Dropping hard roles. A team that stops working the engineering requisitions will post a better overall rate.
Pair the acceptance rate with the number of offers made and with quality of hire for the same cohort, so a rising rate cannot hide a shrinking or weaker set of hires.
Agencies: two acceptance rates
An agency sees offers its clients make to its candidates. The formula is the same, but the agency can also track the rate at which clients' offers to its candidates are accepted by client, which is often the more revealing split. A client whose offers are declined repeatedly may be paying below market, moving slowly, or selling the role poorly at interview. That is a conversation for the account manager, supported by the decline reasons. The recruiting agency metrics page places offer acceptance and fall-off in the full agency funnel.
Calculating it from your ATS
OFFERS (one row per candidate per requisition)
OfferID | ReqID | RoleFamily | Recruiter | OfferExtended |
Outcome (accepted / declined / rescinded / pending) | OutcomeDate |
DeclineReason | Reneged (Y/N) | Internal (Y/N)
Acceptance rate for Q3 (Excel 365, Google Sheets)
=COUNTIFS(Outcome,"accepted",OutcomeQuarter,"2026-Q3",Internal,"N")
/ (COUNTIFS(Outcome,"accepted",OutcomeQuarter,"2026-Q3",Internal,"N")
+ COUNTIFS(Outcome,"declined",OutcomeQuarter,"2026-Q3",Internal,"N"))OutcomeQuarter is a helper column holding the quarter of OutcomeDate as text. Because the formula uses the outcome date, an offer extended in late September and accepted in October counts in Q4, which is the rule set out above.
How to report it
Quarterly is the right frequency for most teams; monthly numbers on fewer than twenty offers swing too much to read. A compact report:
OFFER ACCEPTANCE, Q3
Rule: accepted ÷ (accepted + declined), by outcome date, external
offers only, one offer per candidate per requisition.
Resolved offers: 43 Accepted: 36 Rate: 83.7% (Q2: 86.1%)
Rescinded by us: 2 Pending at quarter end: 3
Reneged after accepting: 2 (5.6% of acceptances)
By role family: Support 17/18, Sales 8/10, Engineering 7/11, Finance 4/4
Decline reasons: compensation 3, counteroffer 2, other offer 1, location 1
Action: engineering pay expectations rechecked at final round from Oct 1.End with one action tied to the largest decline reason. A report that lists reasons without a change attached will show the same reasons next quarter.
Common mistakes
| Mistake | Effect | Fix |
|---|---|---|
| Pending offers in the denominator | Rate understated, then revised later | Count by outcome date only |
| Rescinded offers counted as declines | Candidates blamed for the employer's decision | Report rescinded separately |
| Each revised offer counted | Negotiations look like declines | One offer per candidate per requisition |
| Reneges hidden inside declines | Offer and notice-period problems mixed | Accepted in this metric, plus a renege rate |
| Internal and external offers blended | Internal moves inflate the rate | Split or exclude internal offers |
| No decline reasons | The rate moves and nobody knows why | Fixed categories, recorded the same day |
Questions people ask
How do you calculate offer acceptance rate?
Divide the number of offers accepted by the number of offers that reached a final answer in the same period, then multiply by 100. Leave out offers that are still pending and offers the employer withdrew, and count each candidate once even if the offer was revised.
What is a good offer acceptance rate?
There is no reliable public figure that applies across roles, markets and definitions, so compare against your own trailing rate by role family. A drop of several points over two quarters matters more than where the number sits against someone else's report.
Does a verbal offer count?
Count whichever offer you use consistently. Many teams extend a verbal offer first and send the written one only after a yes; in that case the verbal offer is the real decision point and should be the one counted. Write the rule down and apply it to every role.
Is a candidate who accepts and then backs out counted as an acceptance?
Count them as accepted in the offer acceptance rate, because they did say yes, and track them separately as a renege or fall-off. Reporting both numbers side by side shows whether the problem is at the offer or in the weeks before the start date.