Ask a room full of sales leaders how long a demo should run and you will get two answers, each said with confidence, and both are usually wrong. Half think shorter is always tighter. The other half book a 60-minute slot and fill every minute because the calendar told them to.
Mystery Demo books competitors’ demos as a credible buyer and keeps the stopwatch running: when the product finally appears, how long the rep talks before anyone else can, and the minute the energy in the room quietly dies, then hands the client every one of those times on one board.
We collected the most useful SaaS demo length statistics we could source, drawn almost entirely from recorded sales calls, each footnoted to its original study. The short version: winning demos run longer because they turn into a conversation, and the clock only reports what happened in the room. If you only keep a handful of these, keep these:
Winning Demos Run 47 Minutes, Losing Ones 36
Start with the headline number, because it is the one everyone quotes and almost everyone misreads. In Gong’s 2017 analysis of 67,149 recorded demos matched to deal outcomes, successful demos ran 47 minutes against 36 for unsuccessful ones, a 30.5% gap1.
The catch is that one average hides the size of the deal behind it. Demo length scales with what you are buying, so a single benchmark is close to useless without context. Demodesk’s 2025 reference points from its own customer data show the spread:
| Segment | Typical demo length |
|---|---|
| SMB | 25 to 30 minutes8 |
| SaaS mid-market | 40 to 45 minutes8 |
| Enterprise | 45 to 60 minutes8 |
Put the headline and the segment table together and the 47-minute figure stops being a target and becomes a clue. A demo that runs long because an enterprise buyer keeps asking is healthy.
A demo that runs long because the rep will not stop talking is the same forty-seven minutes spent badly. The number tells you nothing until you know why the clock moved, which is why we watch the call instead of timing it.
Making a Demo Longer Does Nothing on Its Own
Across 30,000 first calls in 2017, Gong found no statistically significant correlation between how long a call ran and the odds of earning a second meeting4. Length on its own did not move the next step.
So winning demos run longer, yet on first calls the extra minutes alone did not earn the next meeting. What connects the two is interaction: a demo earns its extra minutes by provoking questions, objections, and tangents from the buyer.
Padding it with more slides earns you nothing but a politely glazed buyer. You cannot fake the length that wins, because it is a side effect of a conversation the buyer wanted to keep having.
The Monologue Is Where Demos Die
A demo dies in the longest stretch where only the rep is talking, and in Gong’s 2017 data that stretch had a visible ceiling.
No demo that ended in a closed deal contained more than 76 seconds of uninterrupted pitching1, while losing demos regularly ran pitches up to 106 seconds1. Talk about your own company past the 2-minute mark and win rates fall off sharply5.
The product is the thing the buyer showed up to see. Plenty of reps still make them sit through a company-history slide first, as if anyone books a software demo for the founding story.
.png)
The data is blunt about which clock matters: the length of any one stretch where only you are talking. Keep every monologue short enough for the buyer to come back in, and the total runtime takes care of itself.
The Talk Ratio Barely Separates Winners From Losers
A demo is the one call where the rep is supposed to dominate the airtime, which trips up anyone who memorized the standard advice. The winning talk-to-listen ratio for a demo was 65:35, close to the mirror image of the 46:54 that wins a discovery call1. Discovery is for listening and diagnosing, a demo for showing and explaining.
Listen too much in a demo and you leave the buyer waiting for the thing they came to see. Read the ratio alone and you miss the trap underneath it.
| Measure | Won | Lost | Measured on |
|---|---|---|---|
| Demo talk-to-listen ratio | 65:35 | 66:341 | 67,149 demos (Gong, 2017) |
| Questions the seller asked per call | 15 to 16 | about 202 | 326K calls (Gong, 2025) |
| Speaker switches per minute | 21% more1 | the baseline | 67,149 demos (Gong, 2017) |
In Gong’s 2025 analysis of 326K sales calls of at least 10 minutes, not demos alone, the average rep talked 60% of the time, closed-won reps 57% and lost-deal reps 62%2. Top performers held about the same ratio whether they won or lost, while low performers swung from 54% in won deals to 64% in lost ones2.
Two of the 2025 findings rewrite the usual coaching:
What that looks like in the recording is back-and-forth. In winning demos, speaker switches per minute climb 36% in the second half of the call1, and the best reps draw 28% more questions from buyers while asking 30% fewer of their own7.
This is the texture a transcript captures and a stopwatch never will. A demo that gets more interactive as it goes is one the buyer is leaning into, and one that flatlines into a steady drone is one the buyer has already left, whatever the invite says. When we grade a competitor’s demo, the speaker-switch rate in the back half tells us more than the runtime ever could.
The Shape of the Call Beats Its Length
What separates a 47-minute winning demo from a 36-minute loser is where the minutes land. The measured shape of a demo that closes is surprisingly consistent.
| Moment | Calls that closed | Calls that did not | Measured on |
|---|---|---|---|
| Context before the product | under 2 minutes1 | not reported | 67,149 demos (2017) |
| Slide deck in the intro meeting | 9.1 minutes | 11.4 minutes3 | Gong intro-meeting data (2021) |
| When pricing comes up | minute 38 to 461 | no consistent window | 67,149 demos (2017) |
| Time spent talking price | the baseline | 8% more1 | 67,149 demos (2017) |
| Next steps at the close | 12.7% more time, about 4 minutes1 | the baseline | 67,149 demos (2017) |
Read those together and the shape of a winning demo is clear: the product comes fast, the deck stays short, pricing waits until value is built, late in the call, and real time goes to what happens next.
The price details cut against instinct. Spending more minutes on price goes with losing rather than thoroughness, and Gong’s own reading is that the pricing conversation in those demos is less smooth, often because the model itself confuses the buyer.
The healthiest sign is a buyer who asks three or four sharp pricing questions, gets three or four clean answers, and moves on. In Gong’s 2017 data, 3 to 4 buyer pricing questions went with the highest win rates, and fewer or more with declining ones6.
What a Rival’s Demo Clock Shows
All of this is observable from inside the call, which is the point. You will never see a rival’s pipeline review, but you can take their demo the way their buyers do, and the way a rival paces a demo shows how mature its sales motion is.
What no benchmark gives you is the competitor in the next tab: which slide they open on, what they refuse to show, and where their reps lose the room. Our competitor sales tactics research times each of those moments, call by call, on one board. That gives your enablement team a rival’s pacing to coach against: how long their rep talks before the product appears and when price comes up, set beside your own calls.
Below is one competitor’s findings section from our public example board, where the client and every competitor are invented. Its How They Sell notes describe the demo itself, from who ran it to what was shown live.
So how long does your closest rival talk before the product appears? Get us into their next demo, and you will have the answer with a stopwatch on it.
Frequently Asked Questions
How long should a SaaS demo be?
Long enough to solve the buyer’s problem and not a minute longer. In Gong’s 2017 study, demos that closed averaged 47 minutes against 36 for those that did not1, and the length followed the conversation rather than causing it.
How long are demos by deal size?
25 to 30 minutes for SMB, 40 to 45 for SaaS mid-market and 45 to 60 for enterprise, in Demodesk’s customer data8.
What is the right talk-to-listen ratio for a demo?
65:35 in the rep’s favor in winning demos, close to the reverse of the 46:54 that wins a discovery call1.
How long can a rep pitch without a pause?
76 seconds was the ceiling: no demo that closed contained a longer uninterrupted pitch1, while losing demos ran pitches up to 106 seconds1.
How long should a rep talk about their own company?
2 minutes or less: past that mark, win rates fall off sharply5.
How long should the slide deck run?
9.1 minutes in intro meetings that became closed-won deals, against 11.4 minutes in losses3.
When should pricing come up in a demo?
Late, after value is built: in winning demos it came up at minute 38 to 46, while losing demos had no consistent window1.
How many pricing questions should a buyer ask?
3 to 4 buyer pricing questions go with the highest win rates; fewer or more, and win rates decline6.
What is the average talk-to-listen ratio on sales calls?
60% talking and 40% listening across 326K calls in 2025, with closed-won reps at 57% and lost-deal reps at 62%2.
How much time should go to next steps?
12.7% more time in winning demos than in losing ones, about four minutes of logistics at the close1.
References
- Gong Labs: Effective Strategies for Conducting Successful Sales Demos (2017)
- Gong Labs: Mastering the Talk-to-Listen Ratio in Sales Calls (2025)
- Gong Labs: 30 Mind-Blowing Sales Stats (2021)
- Gong Labs: Determining the Ideal Duration of a First Sales Call (2017)
- Gong Labs: Effective Strategies for Discussing Your Company in Sales (2016)
- Gong Labs: Mastering Winning Sales Conversations (2017)
- Gong Labs: How to Know if Your Sales Demo Is Going Well (2018)
- Demodesk: Sales Demo Analysis Made Simple (2025)
.png)
