B2B Win Rate Benchmarks by Deal Size and Segment

The average business-to-business (B2B) sales win rate is about 21% across all opportunities created, and about 29% when only qualified opportunities count. Rates fall as deal size rises: one study of 939 companies put deals under $10K annual contract value at 28% to 35% and deals over $100K at 12% to 18%.

What is the average B2B sales win rate?

The average B2B sales team wins about 21% of its opportunities. Counting qualified opportunities only, the figure rises to about 29%. Both numbers come from the same summary: the average B2B win rate is 21% across all opportunities, rising to ~29% for qualified opportunities only. The 21% traces to HubSpot's survey of over 1,000 sales reps.

Ask a sales leader for their own number and the answer usually runs higher. Leaders quote a number between 30% and 50%. None of these figures contradict each other. They start counting at different points in the funnel.

Where the count starts

The 21% figure typically reflects all pipeline entries, while the 47% figure typically reflects post-proposal conversion. The same source reads the gap between 21% and 29% this way: that 8-point gap represents deals that should never have entered the pipeline.

The definition of an opportunity moves the result too. Some teams count every demo booked; others count only deals that reach a proposal/contract stage. The treatment of stalled deals matters as well: excluding "no decision" outcomes inflates your number by 10-15 percentage points.

One team, three win rates

Take a team with 20 wins. If you won 20 deals and lost 60, your win rate is 25%. Count every opportunity created and the same 20 wins against 150 total opportunities created gives you 13.3%. Now suppose 40 of those deals reached a proposal stage. Against that denominator, 20 divided by 40 gives 50%.

Same team, same quarter, same wins. The reported rate runs from 13.3% to 50% depending on the denominator.

Before you compare your number to any benchmark, write down three things:

1. The stage where your count starts: created, sales qualified, or proposal.
2. Whether no-decision and stalled deals sit in the denominator.
3. The period and whether open deals are excluded.

Then compare against the benchmark row that uses the same starting point. In our reading, a team measuring from proposal stage that reports 35% is not outperforming a team at 21% measured from creation. The two numbers describe different parts of the funnel. If you feed win rate into pipeline coverage targets, use the version that matches the stage where your coverage is counted.

Win rate benchmarks by deal size and segment

The table below separates two kinds of number. Full-funnel win rates count every opportunity in a band. Demo-to-close rates start counting only after a completed product demo, so their denominator is smaller and later in the funnel. Compare your own figure only with rows that start at the same stage.

Measured from completed demos, the reported close rate falls from 32% in SMB to 25% in mid-market and 18% in enterprise.

Demo-to-close rate by segment

Item	Value
SMB	32%
Mid-market	25%
Enterprise	18%

Source: Puppydog, Demo to Close Rate benchmark, 2026

CRM stands for customer relationship management. The Optifai ranges represent the 25th to 75th percentile, so a team at the bottom of a band is still inside the middle half of its peers. Puppydog defines its metric as closed-won deals over completed demos, and it separates this from basic win rate, which covers the full opportunity lifecycle.

Two cautions from our reading. Optifai and Puppydog report the same company count and the same four quarters, so we would not treat them as independent confirmation of each other. And the Trellus ranges are a summary of other reports, so they set a hypothesis, not a target.

Worked example: rate versus revenue

Optifai published a model that shows why a low enterprise rate does not mean a weak segment. SMB: 200 opportunities at 30% and $5,000 average deal produce $300,000. Enterprise: 15 opportunities at 15% and $150,000 average deal produce $337,500.

Check the math yourself: 200 × 0.30 × $5,000 = $300,000, and 15 × 0.15 × $150,000 = $337,500. Enterprise won more revenue from fewer opportunities.

To plan pipeline from these bands, take the rate for your ACV band, not a blended average, and run it through the pipeline coverage calculator. Stage-by-stage rates sit in our funnel conversion benchmarks.

Where the benchmark figures come from and what each one measures

A win rate figure is only as useful as its denominator. Before you set your own number beside a published one, write down who collected it, from how many records, in what year and what it counted. The table below does that for the deal datasets behind the Ebsta figures, plus one marketing efficiency set that often sits next to them in planning decks.

The two Ebsta editions measure different populations. The 2024 report considers small, medium, and enterprise sales processes and uses sales velocity as our ultimate measure. The 2025 edition was produced in association with Pavilion and counts high performers only. In our reading, a figure from that edition belongs in the stretch column of your plan, not the baseline column.

The slippage finding, stated with its baseline

The 2024 Ebsta report found that when deals slipped, win rates plummeted by -67%, especially for deals delayed more than eight weeks. The period is the 2023 selling year the report describes. We read the -67% as a relative drop measured against deals that held their close dates, not a fall in percentage points. A team winning 30% of on-time deals would, on that reading, win about 10% of slipped ones.

If you run pipeline velocity math in our calculator, test slipped and on-time deals as separate inputs before you blend them.

The four-line card

Attach this to every benchmark you quote, including the ACV-band and compiled ranges cited earlier in this piece:

1. Publisher and year: who collected it and which year of deals it covers.
2. Sample and unit: count of companies, opportunities or reps, and total value if given.
3. Win rate definition: what sits in the numerator and denominator, and how no-decision deals are treated.
4. Population fit: segment, deal size and whether the sample is all companies or a selected top group.

If you cannot fill line three, keep the figure out of your plan.

How do you calculate win rate?

The standard formula is short. MetricHQ defines opportunity win rate as dividing won opportunities by all opportunities that reached a final decision, won plus lost, expressed as a percentage.

Win rate = won ÷ (won + lost) × 100

The worked line: 15 opportunities won and 10 lost in a quarter gives 60%, because 15 divided by 25 closed opportunities is 0.6. Open deals never enter the math. Tenbound states it directly: everything still open stays out of the denominator until it closes.

The formula is the easy part. Three choices around it decide what the number means.

Closed cohort or period-closed

A closed cohort takes every opportunity that entered a defined stage in one period, then waits for that group to resolve. A period-closed view divides this quarter's wins by everything that closed this quarter. Tenbound notes the second is faster to read, but mixes cohorts of different ages and qualification standards.

Tenbound's worked cohort: 120 opportunities entered the qualified stage in Q1, and two quarters later 30 had been won, 70 lost and 20 were still open. The cohort rate so far is 30 ÷ (30 + 70), or 30%. The 20 open deals wait. If they resolve mostly as losses, the final rate falls below 30%.

Count-based or dollar-based

Count-based divides deals won by deals closed. Dollar-based divides value won by value closed. In the same cohort, the 30 wins carry $900,000 of $4,000,000 closed value, a dollar win rate of 22.5%. The gap can be wide: a team can win 40% of deals and 15% of dollars when large deals keep dying. Report both, on separate lines.

The stage-entry gate

Where an opportunity starts counting changes the result. MetricHQ's example is a 70% win rate for final-stage opportunities but only 30% for early-stage qualified leads. Tenbound adds that moving the entry gate later raises the reported rate with zero change in selling performance.

A template for your own reporting header, which we use as a checklist: entry stage, cohort period, count or dollar basis, how no-decision outcomes are treated, and the date the definition last changed. When qualification rules change, Tenbound's guidance is to annotate the series and restart the baseline. Win rate then feeds pipeline velocity, which you can model in the pipeline velocity calculator.

Should no-decision deals count as losses?

Yes, in the sense that a no-decision deal is not a win. It should not be filed as a competitive loss either. In our reading, how you treat it moves the reported number more than any other definition choice. Salesmotion puts the size of the effect plainly: excluding "no decision" outcomes inflates your number by 10-15 percentage points.

The stakes are higher in enterprise. The same source says deals that died from inaction, which according to multiple studies represent 40-60% of enterprise pipeline, drop out of view when the denominator holds only competitive outcomes. Tenbound lists the same move among its definition pitfalls: excluding no-decision outcomes so the denominator only holds competitive losses.

A worked example

Take a resolved cohort of 100 opportunities: 25 won, 45 lost to a competitor or another option, and 30 that ended with no decision. These numbers are illustrative.

The gap between the first two rows is 10.7 points, inside the range Salesmotion describes. Same team, same quarter, same deals. Only the denominator changed.

The house view

We track no-decision as its own outcome, separate from closed-won and closed-lost. Then we report win rate both ways, side by side, every period.

Each number answers a different question. The competitive rate tells you how often you beat the alternative when a buyer chooses. The decision-inclusive rate tells you how much of the qualified pipeline turns into revenue. The no-decision rate tells you how often buyers stall, which in our reading points to discovery, urgency and champion strength more than to pricing or product.

Setting it up in the CRM

1. Add a third closed status: closed-no-decision. Keep closed-lost for deals where the buyer picked a competitor, built in-house or bought something else.
2. Write a stall rule. A deal that passes its expected close date by a documented number of days with no buyer activity gets reviewed and either revived or closed as no-decision.
3. Report three lines on one dashboard: competitive win rate, decision-inclusive win rate, no-decision rate.
4. Feed the decision-inclusive rate into pipeline velocity and coverage math, since that is the rate your pipeline actually converts at.

The stall rule matters as much as the status. Tenbound flags the habit of letting reps mark dead deals as open forever, which quietly inflates the reported rate. An open deal that is really dead hides from both denominators.

When you compare against a published benchmark, check which treatment it uses before reading the gap. Salesmotion's advice applies here: Choose one method and stick with it.

One team's win rate under three definitions

Take one illustrative year of a single team's closed opportunities, 200 in total. Of those, 40 were won, 70 were lost to a competitor, 60 ended in no decision and 30 were disqualified at the first stage. Nothing about how the team sold changes in the rows below. Only the denominator does.

The same 40 wins produce a win rate of 20.0%, 23.5% or 36.4% depending on which closed opportunities stay in the denominator.

Worked example: one team's win rate under three definitions

Item	Value
All closed opportunities	20 % won
Excluding early disqualifications	23.5 % won
Won versus lost to a competitor	36.4 % won

Example year: 40 won, 70 lost to a competitor, 60 no decision, 30 disqualified at the first stage. Source: Take the Market worked example (illustrative year of 200 closed opportunities)

The first row divides 40 wins by every opportunity that closed. The second removes the 30 first-stage disqualifications, which is the same move as starting the count at a later stage. The third also removes the 60 no-decision outcomes, so only competitive losses remain in the denominator.

The team won the same 40 deals in every row. The reported rate still comes close to doubling between the first row and the third. In our reading, that spread is larger than the change most teams could produce in a year by selling better, which is why a benchmark comparison means little until the definitions match.

How to report it so the number holds

1. Name the entry stage in writing. State which stage puts an opportunity in the count, and keep it fixed across quarters.
2. Name what counts as a loss. Decide in advance whether no decision and disqualification sit in the denominator. The earlier section on no-decision deals covers why excluding them flatters the rate.
3. Show all three rows once. When you compare against an outside benchmark, publish the full table for one period so readers can line up your definition with theirs.
4. Split by segment and source before you trend it.

What the excluded rows tell you

Dropping the 90 disqualified and no-decision opportunities hides where the losses happened. Tenbound suggests diagnosing win rate by loss reason and stage, noting that losses at the qualified stage point to targeting while losses at proposal point to message or proof. In this example, the 30 first-stage disqualifications are a targeting question for marketing, and the 60 no-decisions are a question about urgency in the pipeline.

Win rate also feeds pipeline velocity. Tenbound describes it as one of the four inputs to that measure, so whichever definition you choose flows straight into the pipeline velocity calculator. Use the definition that matches how the other inputs were counted.

How does win rate affect pipeline coverage and forecasting?

Coverage is the win rate turned upside down. If your win rate is 20%, each dollar of target needs five dollars of qualified pipeline, so coverage is 1 divided by the win rate. The pipeline coverage number is only as good as the win rate behind it, which is why the definition choices in the earlier sections matter here.

Coverage from win rate

Use a win rate measured from the same stage where you count pipeline. Trellus lists the stage-based win rate as "recommended for repeatable forecasting". A late-stage win rate applied to early-stage pipeline overstates the forecast. You can run your own figures in the pipeline coverage calculator.

The 20% to 25% arithmetic

Trellus states that if you increase a 20% win rate to 25% while keeping the pipeline constant, revenue rises by 25%. The math is a ratio of rates.

Five percentage points equal a 25% relative gain. The same move cuts required coverage from 5x to 4x. That frees one fifth of the pipeline you would otherwise need to source. This holds only if deal size stays flat. If the extra wins skew small, check the dollar win rate covered earlier.

Win rate and the new CAC ratio

Benchmarkit defines the new customer acquisition cost (CAC) ratio as Total Sales & Marketing expenses / New Customer ARR. It reports that the ratio increased by 14% in 2024 to a median of $2.00 of Sales and Marketing expense to acquire $1.00 of New Customer ARR. For the fourth quartile, the median was $2.82 to acquire $1.00 of New Customer ARR.

In our reading, win rate is one of the few inputs that improves the CAC ratio without cutting spend, since it changes the denominator. The spend that built the pipeline is already sunk.

For the payback view of the same spend, see CAC payback and our CAC payback benchmarks.

How to benchmark your own win rate in six steps

This template turns the definitions from earlier sections into a repeatable report. Run it once a quarter, and change it only through step six.

1. Fix the entry stage. Write down the one customer relationship management (CRM) stage that admits an opportunity to the cohort, such as "qualified evaluation," and stamp every opportunity with the date it entered that stage.
2. Build closed cohorts. Group opportunities by entry quarter. Report a cohort's rate only once most of it has resolved, and print the share still open next to the rate.
3. Split by ACV band. Optifai's study used four bands: <$10K, $10K-$50K, $50K-$100K and >$100K. Its published ranges represent 25th-75th percentile, so compare your band rate to a range, not a single point.
4. Report count and dollar rates side by side. One column each, never blended in a single trend line.
5. Log no-decision separately. Use three closed outcomes: won, lost to a competitor, no decision. Publish the rate with and without no-decision so readers see both.
6. Annotate definition changes. Keep a changelog row with the date, what changed and which cohorts are affected. A new definition starts a new series.

Worked template row

The figures below are illustrative, for one cohort in one band.

The 8 open deals sit outside every denominator until they close. Feed the count rate into the pipeline velocity calculator and add the table to your demand gen dashboard.

Add the driver fields

Two more columns turn the report from a score into a diagnosis, because the reports measured both drivers.

• Decision-maker engaged early (yes or no). Ebsta and Pavilion reported that early decision-maker involvement boosts win rates by 55%. Define "early" as before a named stage and hold that definition the same way you hold the entry stage.
• Close date pushed (count of pushes). The same report found delayed deals reduce win rates by 113%. Count each change to the expected close date, and compare the win rate for deals with zero pushes to deals with one or more.

Optifai also measured stakeholder breadth: deals with 3+ contacts engaged close at 2.4x the rate of single-threaded deals. A contact-count column costs little to add. Slippage also stretches cycle time, so read this report next to your sales cycle length benchmarks.

What this page does not cover

This page does not cover stage-by-stage funnel conversion, sales cycle length or CAC payback in depth. Those topics sit in our separate benchmark pages. It does not rank or recommend any benchmark publisher or tool, and nothing here is financial or legal advice.

Related on Take the Market: MQL to SQL conversion rate benchmarks and B2B funnel conversion rates by stage; Average B2B Sales Cycle Length by Deal Size and Segment; Demand generation calculator: reverse funnel math from a revenue target to visitors, leads, MQLs, SQLs and opportuniti.