Demo-to-Close Rate: What the Benchmarks Measure, How to Calculate Yours, and How to Raise It
Demo-to-close rate is the share of product demos that become closed-won deals within a set window, usually 90 or 180 days. Published figures for B2B SaaS mostly land between 15% and 35%, but they count different demos over different windows, so comparing them is harder than it looks. This guide shows what each benchmark measures, how to calculate your own rate by cohort, and the six changes most likely to raise it.
One quarter of demos
18%
of booked
23%
of held
35%
of opportunities
Same 18 wins, three honest-looking demo-to-close rates.
What Is Demo-to-Close Rate?
Demo-to-close rate is the share of product demos held in a period that become closed-won deals within a fixed window after the demo. Formula: won deals from a cohort of demos ÷ demos held in that cohort, measured at 90 or 180 days.
It answers one question: when we give a demo, how often does it turn into revenue? That makes it the number to watch when you change who gives demos, how they run them, or which leads get one. It is a narrower number than win rate, which starts from opportunities rather than demos, and a later one than show rate, which our guide to demo no-shows covers.
The pages that rank for this term mostly repeat the same handful of figures, often without saying where they came from or what was counted. This guide starts with the definition, because most of the disagreement between benchmarks comes from there.
One disclosure: SnapSDR is an AI demo agent that runs on Tough Tongue AI's agent infrastructure. It comes up once, in the section on separating first demos from later ones, as one of several ways to do that.
How Do You Calculate Demo-to-Close Rate?
You make three choices, and each one moves the number.
Which demos count. Count first demos that were held, not booked. Leave out demos to existing customers, partners, and internal teams, or track them on their own line. If a buyer sees three demos, the deal counts once, against the first.
Which deals count. Only closed-won deals that came from that cohort of demos. Not every deal won this month.
How long you wait. Pick a window that fits your sales cycle: 90 days for most SMB and mid-market products, 180 days when deals above $50,000 are common. Report it next to the number, every time.
The most common mistake is dividing this month's wins by this month's demos. Most of this month's wins came from demos held two or three months ago, so that ratio rises and falls with pipeline timing, not demo quality. Use a cohort instead: every demo held in March, and what became of each one.
One Funnel, Six Different Rates
Here is why two teams with the same results can report very different rates. Pick a denominator and a window.
One quarter of demos at one company. Change what you divide by and how long you wait, and watch the rate move.
Divide by
Count wins within
13 won ÷ 78 demos held =
16.7%
Counts every demo that actually happened. This is the definition we recommend, and the one closest to Optifai's “demos delivered”.
Illustration. The funnel numbers are invented to show how the definition changes the result. The same quarter reads anywhere from 13% to 35%.
Neither choice is wrong, as long as you say which one you made and keep it fixed. Divide by demos held, count wins within a window that fits your cycle, and track demos that never became an opportunity as their own number, because that is where most of the fixable loss sits.
What Is a Good Demo-to-Close Rate?
We read the pages that rank for this term in November 2026. Four sources come up again and again, and they do not measure the same thing. Only two of them say how many companies their demo-to-close figures come from.
| Source | Who publishes it | Sample | Window | Headline figure |
|---|---|---|---|---|
| Optifai | Sales pipeline software vendor | 939 B2B companies, Q2 2025 to Q1 2026, from CRM, demo-platform, and industry-report data | 90 days after the demo | 25% average; 30% for B2B SaaS |
| PresalesPulse | Career and salary site for sales engineers | Practitioner survey, PreSales Collective groups, CRM data from 38 B2B SaaS companies | 180 days | Medians from 15% (scripted) to 41% (POC-anchored) |
| Naoma | AI demo agent vendor | Not stated | 90 days in one figure, not stated elsewhere | Median 15% to 20% in its funnel table; 25% in its FAQ |
| Ebsta and Pavilion | Sales analytics vendor and a sales community | 4.2 million opportunities, 530 companies, $54 billion in revenue | Deals through 2023 | No demo-to-close figure; win rates down 18% on 2022 and 27% on 2021 |
Optifai is the most specific: it names its sample, its time period, and its formula (closed-won deals divided by demos delivered, at 90 days). PresalesPulse uses a longer 180-day window and mixes three inputs without saying which produced which number. Naoma gives no source or sample for its close-rate figures. Ebsta and Pavilion is the largest dataset, but it reports win rates rather than demo-to-close rates.
Switch between the cuts below. The pattern holds across sources even where the levels differ: smaller deals close at higher rates, and demos built around the buyer's own problem close at higher rates than standard tours.
Published demo-to-close rates, by how the data is cut
15 to 30 day cycle
30 to 60 day cycle
60 to 90 day cycle
90 to 180 day cycle
Optifai, 939 B2B companies, Q2 2025 to Q1 2026, closed-won deals divided by demos delivered at 90 days. Note the cycle lengths: most $100K deals cannot close inside a 90-day window. Read November 2026.
Why the Published Numbers Disagree
Three reasons, in order of size. First, the window. Optifai measures at 90 days, but its own data puts the sales cycle for deals above $100,000 at 90 to 180 days, so many of those deals have not had time to close when the count is taken. Part of the enterprise gap is the window, not the demo.
Second, the denominator, as the figure above shows. A source that counts qualified opportunities will report a higher rate than one that counts every demo held.
Third, the sample. Vendor datasets come from that vendor's customers, who already bought a sales tool and are likely to run a tighter process than average. Our advice: treat 15% to 35% as the range most B2B SaaS teams fall in, expect the low end for large deals and the high end for small ones, and use your own rate from last year, measured the same way, as the benchmark that matters.
Why Is Your Demo-to-Close Rate Low?
Split the rate in two before you fix anything. Demo to opportunity tells you whether you are demoing the right people. Opportunity to close tells you whether the demo and what follows it are working. A team with a low first number needs better qualification; a team with a low second number needs better demos and follow-through. Then find where the deals stop.
| Where it breaks | What you see | What to check |
|---|---|---|
| Before the demo | Many demos never become an opportunity | How many demos were booked straight from a form with no questions asked, and how many had discovery first |
| In the demo | Buyers go quiet afterwards | Whether the demo followed the buyer's use case or the standard tour, and how many people from the buyer's side attended |
| Right after | Deals sit for weeks with no meeting | The share of demos that ended with a dated next step, and the days from demo to the next meeting |
| Late stage | Losses to “no decision” | Whether the buyer had a business case in their own numbers before pricing came up |
| Late stage | Losses to a competitor on features | Whether deals that needed a technical demo or a proof of concept got one |
Most of these checks need you to listen to demo recordings, not just read the CRM. A short sales call scorecard for demo calls (did discovery happen, did the demo follow the buyer's use case, was a dated next step agreed) turns that listening into numbers you can compare across reps and months.
How Do You Raise Your Demo-to-Close Rate?
Six changes, roughly in order of cost. The figures quoted are correlations from the sources above, not controlled experiments, so test each change on part of your pipeline before you roll it out.
Qualify before you demo
A demo for a lead who will never buy costs the full hour and lowers the rate. Ask three or four questions first, in the form, a chat, a call, or the first minutes of the demo.
Run discovery first
PresalesPulse's medians put discovery-led demos at 28% against 15% for scripted ones. Even ten minutes of questions before the screen share changes what you show.
Show their workflow, not your menu
Optifai reports 25% for customized demos against 18% for a generic screen share. Build the demo around the two or three jobs the buyer named in discovery.
Get more of the buying group on the call
Multi-stakeholder demos sit at 31% in Optifai's data. Ask who else should see it when you book, and send a short recap they can forward when they cannot attend.
End every demo with a dated next step
Ebsta and Pavilion found win rates fell 67% when deals slipped, especially past eight weeks. A meeting on the calendar before the call ends is the cheapest fix on this list.
Split first demos from evaluation demos
The first demo answers “is this for us?”. Later demos answer “will this work for us?”. Give the second kind your best SE time, and let the first kind run faster.
On the first change, our guide to AI lead qualification covers which questions to ask and how to check you are not turning real buyers away. On the third, our product demo best practices go through how to structure a demo around the buyer's problem.
The sixth change is the one most teams skip. When reps and SEs give every first demo themselves, evaluation demos get whatever time is left, which is how the SE to AE ratio becomes a bottleneck. You can move first demos to a self-serve interactive demo, a pooled team of junior reps, or an AI demo agent such as SnapSDR that joins the call and demos in a live browser. Whichever you choose, measure that cohort on its own line.
What Is a Few Points Worth?
Put in your own numbers. The second result is often the more useful one: the same revenue from fewer demos is time your reps get back.
What a few points of demo-to-close rate are worth over a year
Extra first-year revenue
$648,000
Or the same deals from
12 fewer demos a month
Simple arithmetic on your own inputs, not a forecast. It assumes the extra wins come from the same deal mix, which is rarely exactly true.
How Should You Track Demo-to-Close Rate?
Use a cohort table: one row per month of demos, and columns for how many of them had closed by 30, 90, and 180 days. Each row fills in over time, and you compare rows only at the same age.
| Demo month | Demos held | Won by 30 days | Won by 90 days | Won by 180 days |
|---|---|---|---|---|
| May | 64 | 4 | 13 (20%) | 17 (27%) |
| June | 58 | 3 | 12 (21%) | Still open |
| July | 71 | 5 | 15 (21%) | Still open |
| August | 66 | 4 | 14 (21%) | Still open |
| September | 60 | 3 | Still open | Still open |
Illustration with invented numbers, read in the first week of December. A cell fills in only once every demo in that month is old enough.
Then cut each row four ways: by lead source (inbound, outbound, partner), by deal size band, by who ran the demo (rep, SE, self-serve, or AI agent), and by demo type (first demo or evaluation demo). A change in the total usually turns out to be a change in the mix, such as more small deals or more outbound demos, and the cuts show it.
Watch the sample size. With 50 demos in a cohort and a true rate of 25%, chance alone can put the measured rate anywhere from about 13% to 37%. Compare rates by rep only over a quarter or more, and do not react to one month.
Frequently Asked Questions
What is demo-to-close rate?
Demo-to-close rate is the share of product demos that become closed-won deals within a set window. The usual formula is the number of won deals from a cohort of demos held in a period, divided by the number of demos in that cohort, measured 90 or 180 days after each demo. Say which demos you count (booked, held, or qualified) and which window you use, because both change the number.
What is a good demo-to-close rate?
Published figures for B2B SaaS mostly fall between about 15% and 35%. Optifai, which sells sales pipeline software, reports a 25% average across 939 B2B companies (30% for B2B SaaS), measured 90 days after the demo. PresalesPulse reports medians from 15% for scripted demos to 41% for demos built around a proof of concept, measured at 180 days. Small deals close at higher rates than large ones. The most reliable benchmark is your own rate from last year, measured the same way.
How do you calculate demo-to-close rate?
Pick a cohort, such as every first demo held in March. Count how many of those deals were closed-won within 90 days of their demo, and divide by the number of demos in the cohort. Do not divide this month's wins by this month's demos: most of this month's wins came from demos held months earlier, so that ratio moves with pipeline timing rather than with demo quality.
Why is my demo-to-close rate low?
Split it into two stages. If many demos never become a real opportunity, you are demoing leads who were not qualified or skipping discovery. If opportunities form but stall, look at next steps, how many people from the buyer's side joined, and how long deals sit after the demo. Ebsta and Pavilion's 2024 B2B Sales Benchmarks found that win rates fell 67% when deals slipped, especially deals delayed more than eight weeks.
What is the difference between demo-to-close rate and win rate?
Win rate usually divides won deals by all closed opportunities, won plus lost, whatever stage they reached. Demo-to-close rate starts from the demo, so its denominator includes demos that never became an opportunity and deals still open when the window ends. For the same team, demo-to-close rate is usually the lower of the two.
Does an AI demo agent change demo-to-close rate?
It changes what you should measure. If an AI demo agent or a self-serve interactive demo handles first demos, track that cohort separately from demos run by reps, and compare the two on won deals per demo request as well as per demo held, because an agent can take every request while reps take only the ones they can staff. Treat vendor claims of higher close rates as claims to test on a share of your own demo requests.
Fix the Definition, Then the Demo
Demo-to-close rate is only useful if it means the same thing every month. Divide won deals by first demos held, count wins within a window that fits your sales cycle, and track by cohort. Compare yourself to your own history before any published benchmark, and treat the 15% to 35% range as a rough guide rather than a target.
Our recommendation: split the rate into demo to opportunity and opportunity to close, fix whichever is lower, and start with the cheap changes: qualify before the demo, run discovery first, and end every demo with a dated next step. When first demos are eating the time your best people need for evaluation demos, move first demos to a self-serve demo or an agent, and measure that cohort on its own.
Take First Demos Off Your Reps' Calendars
SnapSDR joins Google Meet or Zoom, gives the first demo in a live browser, qualifies the buyer as it goes, and hands complex deals to your team with the full context, so your reps spend their hours on the demos that close.
Related Reading
SnapSDR Team
AI Sales Automation Experts