Win Rate Benchmarks for B2B Sales Teams

Several B2B sales benchmarks put win rate somewhere around 20%: HubSpot's survey data gives an average of 21%, and Ebsta and Pavilion arrived at 19% for new business after analysing 655,000 opportunities. Results vary considerably by dataset and by how win rate is defined, though, and there is no single shared "good level". The same Ebsta dataset puts expansion deals with existing customers at 45%, which means deal type shifts the number more than any industry comparison. The benchmark worth having is your own number segmented by rep, deal type, and lead source.

Win rate is one of the most-watched metrics in B2B sales, and one of the most misused. It compresses a complex process into a single number, which makes it easy to report and hard to interpret.

This article covers how to calculate win rate consistently, what ranges are typical at different deal sizes, and how to use the number to diagnose your sales process instead of reporting it every month without drawing conclusions.

How Do You Calculate Win Rate?

Win rate is the share of sales opportunities your team wins. The common formula is straightforward:

Won deals ÷ all closed deals (won + lost) × 100

The definition varies between organizations, though, and that matters more than most managers realize. Some teams count every deal that entered the pipeline, others only deals that reached a proposal. Neither is wrong, but the two numbers are not comparable.

How much the definition moves the number shows clearly in HubSpot's data: the same report gives an average win rate of 21% and an average close rate of 29%. That is the same selling work measured two ways, eight percentage points apart. If your own number doesn't say which metric it is, it shouldn't be compared to anything.

Two things are worth locking down before the number is used for anything:

1. Pick one method and stay with it. If you compare your number to an external benchmark, make sure the denominator matches. Otherwise you are comparing two different things.
2. Clear out the dead pipeline. Deals that were never actively worked drag the win rate down artificially. Define when a deal counts as an opportunity and exclude the rest from the calculation.

This is the first place where data quality decides the outcome. If stage updates depend on a rep remembering to make them, the calculation reflects who kept records, not who won deals.

What Is a Good Win Rate in B2B Sales?

There is no universal good number, but two independent datasets land in the same place. HubSpot reports an average win rate of 21% based on a survey of over 1,400 sales professionals, and Ebsta and Pavilion arrived at 19% for new business after analysing 655,000 opportunities. The methods are opposites, self-reported survey against figures calculated from CRM data, yet the result lands in the same spot. The thresholds below are Empirica's interpretation of what your own number means relative to that roughly 20% level. The boundaries are not from the sources, they are derived from them, and they apply to new business.

Win rate	Interpretation	First action
Below 10%	Serious issue	Review the sales process stage by stage
10–15%	Needs improvement	Improve qualification and lead quality
15–25%	Typical level, benchmark 19%	Work on competitive differentiation
25–40%	Good performance	Spread what works across the whole team
Above 40%	Excellent for new business	Check whether the number includes expansion deals

The single biggest driver isn't industry, it's deal type. In the same dataset, the gap between a new business deal and an expansion sold to an existing customer is roughly twofold:

	New business	Expansion to existing customer
Win rate	18%	45%
Sales cycle length	91 days	52 days
Stakeholders involved	8	5

The report states the new business win rate at slightly different precision in two places, 19% in the overview and 18% in this comparison. That is rounding rather than a contradiction, and the order of magnitude is the same.

Selling to an existing customer is, in the report's own words, about 2.5 times easier, and 52% of new revenue last year came from existing accounts. If your win rate looks unusually good, the first thing to check is how much of it is expansion business.

Lead source matters too, but less than people usually assume. The report measures sources by an efficiency figure that combines win rate, deal size, and cycle length. The order is partner referral 1.3x, organic inbound 1.2x, outbound 1.05x, events 0.78x, and paid 0.68x. Inbound does beat outbound, but calculated from those figures the gap is about 15%, not several times over.

Deal size moves the number in the same direction as stakeholder count: the larger the deal, the longer the cycle and the lower the win rate. It isn't worth hunting for precise benchmarks per deal size, because your own historical data in the same segment is the only benchmark that tells you something about your team.

Why Does Win Rate Alone Mislead You?

A single number hides what it is made of. Two misreadings show up in almost every team.

A high number is not automatically good. If your win rate is 60%, the first question isn't how to keep it there. It's whether you are entering too few competitive situations, and whether expansion business is mixed into the number. Reps may be picking only the safe deals. That looks good in the report and caps growth.

A low number is not automatically bad. 15% sounds alarming, but it is close to the measured 19% level for new business. In large enterprise sales, where a single win covers the losses several times over, even a lower figure can be profitable. Profitability decides, not the percentage.

So the right question isn't "what is our win rate". It's "what is it on the right opportunities, and how far apart are our best and weakest reps".

Which Metrics Should You Track Alongside Win Rate?

Win rate tells you the outcome, not the cause. Four metrics alongside it give you a picture you can act on.

Metric	What it tells you	What it exposes
Pipeline coverage (pipeline value against target)	Whether there is enough raw material to hit the target	A thin pipeline or an unrealistic target
Average deal size	What kind of deals you actually win	The target group drifting toward smaller deals
Sales cycle length	How long the decision takes	An unmapped buying process or a missing decision-maker
Win rate by rep	Whether this is a process issue or a skills issue	Wide variance means the practice was never spread

That last row is the most revealing of the four. In Ebsta's dataset, 14% of sellers generate 80% of revenue, and the sales velocity gap between top and bottom performers is elevenfold. It comes from four factors at once: top performers have a 43% higher win rate, a 42% shorter sales cycle, a 76% higher average deal value, and 2.64x more deals in flight. None of that shows up in the headline number. At the same time 78% of sellers missed quota, up from 69% a year earlier, which makes comparing yourself to the average an even poorer yardstick.

Pipeline coverage is worth looking at for quality, not just value. We covered separately how AI can identify stalled revenue before your team does, well before it shows up as a falling win rate.

What Causes a Low Win Rate?

Five causes come up far more often than the rest. They are not mutually exclusive, and usually at least three are present.

1. Weak qualification. The team works deals that were never winnable. This is the cause we run into most often in our assessments, and the easiest to fix, because it also frees up rep time immediately.
2. An unmapped buying process. Nobody knows who decides and on what grounds. The deal moves along nicely until it stops at an approval step no one knew about.
3. Weak differentiation. The rep can't articulate why this option beats the alternative. It shows up most clearly in lost deals where the recorded reason is price.
4. A mismatch between target group and pipeline. The pipeline holds customers who don't match the ideal customer profile. The number often improves simply by narrowing who you contact.
5. Poor data quality. Stages are out of date, loss reasons go unrecorded, and the same field lives in two systems in different formats. In Ebsta's dataset, 44% of the contacts reps actually engage with never make it into the CRM at all, and about a quarter of those missing contacts are senior decision-makers. In practice that means nobody can report how many decision-makers a deal really involves. This is typically a symptom of manual processes, whose hidden costs run well past the lost hours.

The fifth cause is what blocks fixing the other four. If loss reasons haven't been recorded consistently, causes two and three can't be verified, so they get discussed as opinions.

How Do You Use Win Rate as a Diagnostic Tool?

One number reported monthly steers nothing. Four moves make it useful:

• Segment it. Calculate it separately by rep, deal size, lead source, and industry. The headline number is an average that hides exactly the information you need.
• Look at it stage by stage. Find out where deals die. A deal lost at proposal points to a different problem than one that goes quiet after the first meeting.
• Compare internally. The gap between your best rep and the team average is more useful than the gap to an industry figure.
• Categorize loss reasons systematically. Fixed reason codes and a regular review, not a free-text field nobody reads.
• Track deal slippage. This is the most predictive single signal of the four. In Ebsta's dataset the win rate falls as a deal slips: 18% for deals that slip a week, 13% at a month, 8% at three months, 5% at six months, and 3% beyond that. A late-stage deal that slips by more than two months is effectively a different deal from the one in the forecast.

The practical obstacle is usually effort. When a segmented view is compiled by hand into a spreadsheet, it gets done once a quarter and then forgotten. This is where automation pays off most: compiling the report is repetitive, rule-based, error-prone work, which is exactly the kind worth automating first. Once the same view refreshes itself weekly, win rate becomes a steering signal rather than a line in a monthly report.

The same applies to forecasting. Stage probability should be calculated from historical win rates rather than estimated deal by deal, something we covered in why sales forecasts fail.

Where Should You Start?

If you recognize your own team in this, these steps move things forward without a separate project:

1. Lock down the calculation method and recalculate the number for the last 12 months.
2. Segment it by deal size and lead source. Find the segment dragging the average down.
3. Review the last 20 lost deals and categorize the reasons.
4. Count how many hours a month go into compiling sales reports, and whose hours they are.
5. Automate one manual compiling step and measure the time saved.

If the list won't fall into order on its own, a fixed-price Automation Assessment walks through your sales processes and tells you which targets to automate first and what they're worth in euros. Implementations are built on Microsoft Azure, and we're a Claude Partner Network member, so where data is processed and who can access it is known from the start.

The difference between a 19% and a 30% win rate is rarely talent. It's how strictly you qualify, how reliable the data is, and how quickly a deviation gets noticed.

Frequently Asked Questions

What is a good win rate in B2B sales?

Several benchmarks put the figure around 20%: HubSpot's survey data gives an average of 21%, and in Ebsta and Pavilion's dataset of 655,000 opportunities the new business figure is 19%. For expansion sold to existing customers the same figure is 45%, so deal type matters more than industry. Derived from those, 15–25% is typical for new business, above 25% is good performance, and below 10% usually points to a qualification problem. These boundaries are an interpretation of measured averages, not themselves measured.

How do you calculate win rate?

Divide won deals by all closed deals, meaning won plus lost, and multiply by 100. Some teams include every deal that entered the pipeline, others only those that reached a proposal. Either method works as long as you stay with it and exclude deals that were never actively worked.

Is a high win rate always a good thing?

No. A figure above 40% in new business can mean the team only enters competitive situations it knows it will win. That looks good in the report but caps growth, because too few opportunities enter the pipeline. The other common explanation is that the number includes expansion business with existing customers, where the benchmark is 45%. Check the mix of deal types, average deal size, and pipeline coverage against target at the same time.

Why is our win rate low?

The most common causes are weak qualification, an unmapped buying process, weak competitive differentiation, a mismatch between the target group and the pipeline, and poor data quality. The last one is the hardest, because it prevents verifying the others: if loss reasons aren't recorded consistently, causes get discussed as opinions rather than observations.

---

Empirica helps growth companies remove operational bottlenecks with automation and operational AI. The Automation Assessment is a fixed-price way to find out where to start.

← Back to home