B2B win rate benchmarks 2026: what is normal

A win rate benchmark can serve two purposes: a number to measure yourself against, or a way to understand what moves your own number. The first only works if the benchmark counts the same thing your CRM counts, and hardly any do. The second always works, and it is where this piece ends up.

The number everyone quotes is a survey

'The average B2B win rate is 21 percent' turns up in almost every article on the subject, often attributed to a study of 847 SaaS companies that nobody can locate. The figure matches a pair of HubSpot survey results in which reps report an average win rate of 21 percent and an average close rate of 29 percent. Those are reps estimating themselves, not outcomes measured in a CRM, and the page states no sample size, no year and no definition of how the two measures differ. As a yardstick for your team it is useless. As an example of how a number gets copied until it turns into truth, it could hardly be better.

Three buckets that must not be mixed

• CRM outcome data: Ebsta and Pavilion's benchmark series (4.2 million opportunities from 530 companies in the 2024 report, 655,000 from 349 companies in the 2025 one) and its successor from Fullcast (361,000 opportunities, 2,500 sellers, 78 billion dollars of pipeline in 2026), plus Gong's analyses of recorded deals. UK and US tech companies. Published as year-on-year change, never as an absolute level.
• Company-reported surveys: The Bridge Group's 2026 AE report (158 B2B companies) and RAIN Group's study of 472 sellers and executives. They measure the share of reps at quota, median quota and ramp time, not win rate.
• Reps' own estimates: HubSpot's State of Sales (around 1,000 sales professionals) and Salesforce's State of Sales 2026 (more than 4,000). Perceptions of win rate, cycle length and quota.

The rule is simple: compare a number only with others from the same bucket. Ebsta's decline in win rate cannot be set against HubSpot's 21 percent, and Bridge Group's quota attainment tells you nothing about win rate at all.

The trend in the only comparable series

Ebsta and Pavilion's March 2024 report, built on 4.2 million opportunities, found that win rates fell 18 percent in 2023 compared with 2022 and sat 27 percent below the 2021 level. The 2025 report, covering 48 billion dollars of opportunities from UK and US tech firms, showed the fall slowing: down 18 percent in 2023, down 10 percent in 2024. Over the same period the share of reps missing quota rose from 69 percent in 2023 to 78 percent in 2024, and Fullcast's 2026 benchmark puts it at 78.3 percent. Read the two curves together: win rates are falling more slowly, but quotas have not had time to catch up. Ebsta's own footnote from 2024 is the most important line in the whole series: quotas had been cut by 19 percent, and had they stayed flat, 79 percent of reps would have missed instead of 69.

The Bridge Group measures the same thing from the employer's side. In June 2026 it reported that 48 percent of reps hit annual quota, down from 51 percent in 2024, that median quota is now 960,000 dollars and that ramp time has reached 6.2 months. One detail in the same report is worth noting: companies in the top third of AI use had 57 percent of reps at quota, against 39 percent in the bottom third. That is a correlation, not proof, but it points the same way as Salesforce's finding that 57 percent of sales professionals say the sales cycle is getting longer.

What moves win rate most: qualification

Ebsta's 2025 qualification report is the most useful number in this whole review, because it measures something you can change. Deals with strong qualification scores won 50 percent of the time, poorly qualified ones 8 percent. Well-qualified deals won 6.3 times as often and closed 21.6 percent faster. And only 36 percent of deals that passed discovery had full qualification data. The gap between 50 and 8 percent is not in the product or the market. It is in what the rep found out, and wrote down, in the first two calls.

Gong's data supplies the second lever. In an analysis of 10,332 deals, win rate was 15 percent when the rep sold alone, deals with three or more people on the selling side had more than three times that win rate, and deals with a C-level person involved had the highest win rate of any persona. A later analysis of 184 companies found that win rates doubled when a C-level executive was involved compared with deals that had no executive at all. Timing matters too. Gong's study of more than a million executive sales cycles, published in February 2026, shows win rates dropping by roughly 6 percent when the evaluation starts with the executive, and rising 5 percent when the executive joins around the third touchpoint.

Benchmark against yourself

No Swedish win rate benchmark has been published. We checked Mercuri International, Säljarnas and the Swedish consultancies' sites and found nothing. That leaves one yardstick that means anything: your own, per stage and per rep. Four measures are enough.

1. Define win rate once: won divided by won plus lost, per quarter, and write the definition down so it does not change when the number gets uncomfortable.
2. Measure per stage: share of meetings that become proposals, share of proposals that close. Ebsta's 50 versus 8 percent shows the proposal stage, and what happened before it, is what decides.
3. Measure per decision level: win rate in deals with and without a C-level person involved, and when in the deal that person came in.
4. Measure per rep and compare with the rep's own last quarter, not with the team average. Bridge Group's 48 percent at quota is a distribution, not a target.

Our article on leading and lagging sales metrics covers how to tie the measurement to behaviours you can influence week by week, and the win-loss analysis guide shows how to find out why a deal was actually lost.

Put Ebsta's data in training terms and it says one thing: win rate is decided in discovery and qualification. That is exactly what you practise in Salesprep, against an AI buyer with a hidden goal, a budget owner it will not mention unless you ask, and a decision criterion you have to dig out question by question. Every call is scored on seven metrics with a written comment per score. Cold call is in every plan with three free calls, no card required. Pitch, where the full qualification is practised, is available on Team Pro.