Product Demo Video Length Is Set by Where the Video Plays

The spread between the shortest and longest row is roughly 100 to 1. Any advice that gives you one number is describing one row and calling it a rule.

Runtime is not a creative decision. It is a word count.

The average English speaker talks at about 150 words per minute, and roughly 140 words per minute is the recommended rate when clarity matters most (National Center for Voice and Speech, cited by Baruch College Tools for Clear Speech). Technical voiceover for a product demo sits in the 140 to 160 range. Push past 170 and comprehension falls apart, particularly for viewers reading a UI at the same time.

Then subtract. Take 15 to 20% off those word counts for breathing room, for the two seconds a UI change needs to register, and for the pause before a result lands. A 60 second video with 150 words in it is a 60 second video with no pauses. In practice you write 120 to 135.

This is why "make it shorter" and "add the security section" cannot both be actioned. No editing technique puts 300 words into 60 seconds. Something gets cut, and it is better that you choose than the edit.

Shorter videos finish at higher rates. That is consistent and not in dispute.

• Videos under one minute keep 65% of viewers to the end, while videos over 20 minutes keep 20% (Vidyard, 2025 Video in Business Benchmark Report, based on 943,305 videos created in 2024).
• Videos under a minute average a 52% engagement rate (Wistia, 2026 State of Video Report, 13 million videos analyzed).
• Videos of 3 to 5 minutes average a 43% engagement rate, which rises to 74% for instructional videos in the same range (Wistia, 2024 State of Video Report).
• 71% of people say videos between 30 seconds and 2 minutes are the most effective length (Wyzowl, 2026 State of Video Marketing, 266 respondents).

Now the part that gets misread. Completion rate is a ratio, not a quantity. 65% of a 45 second video is 29 seconds of attention. 20% of a 20 minute video is four minutes of attention. Wistia found that a 30 to 60 minute video has roughly half the engagement rate of a 1 to 3 minute video, and over ten times the total watch time (Wistia, 2024).

Optimizing for completion rate optimizes for shortness. That is the right call when the viewer did not ask for the video. It is the wrong call when they did.

The two minute rule was derived from homepage and social data, where the viewer never requested the video and attention is the scarce resource. In a sales follow up to an evaluator, attention is not scarce. Completeness is.

By the time someone asks you for a detailed demo, they have done most of the work. B2B buyers are nearly 70% through their purchasing process before engaging with sellers, and 85% have largely established their purchase requirements before reaching out (6sense, 2024 B2B Buyer Experience Report). They are not browsing. They have a list.

Send that person a 90 second video and you have not been concise, you have been evasive. They asked whether it handles their SSO and SCIM setup. A 90 second overview answers nothing, so they book another 40 minute call, which is the exact cost the video existed to remove. Wistia's 2026 data points the same way for this sector: in software and technology, videos between 3 and 30 minutes perform best, with audiences typically in research mode.

Two things make long demos work. Chapter them, so a nine minute video with six chapters behaves like six short videos on one URL. And put timestamps in the email body, so the evaluator can go straight to the section they asked about. If the product is an AI tool, the runtime also has to absorb real model latency, which we cover in how to make a demo video for an AI product.

Runtime is not what viewers experience. Density is.

The things that make a demo drag are almost always at the front: a login screen, an empty state, a tour of the navigation, a twelve second logo animation, and a narrator saying "as you can see" before anything has happened. Cutting 40 seconds of setup from a four minute demo makes it feel shorter than trimming it to three minutes with the setup left in.

Three rules we hold to:

• The product does something visible within eight seconds. Not the logo, not the problem statement, the product.
• Edit at two speeds. Move fast through anything the viewer already understands, then slow down and hold at the moment of proof. A demo at one constant pace feels twice its length.
• Every feature shown must have been asked about by a real prospect. Feature tours are long because they are ordered by your product roadmap rather than by the viewer's doubts.

The same discipline applies to the short-form cutdowns you make from the master edit, where six seconds of setup is the whole budget.

If we can only make one video, how long should it be? 75 to 90 seconds. It is long enough to show a full workflow and short enough to sit on a homepage. Then cut a 30 second and a 15 second version from the same master for paid social, and record a longer sales version separately when budget allows.

Does the two minute rule apply to YouTube? No, and following it there hurts you. YouTube ranks on watch time, and a viewer who searched your exact query is prepared to stay. Four to twelve minutes is normal for a product walkthrough, which is why we treat it as long-form work rather than a cutdown.

Should the homepage video autoplay? Muted, looping, and under 60 seconds if it does. Write it to work with no audio at all: burned captions or an on-screen text layer, and a first frame that reads as a still image, because most visitors will only ever see that frame.

How long should the version in onboarding be? 20 to 90 seconds per step, never one long video. Someone stuck on step three will not scrub a six minute tutorial to find it. One video per blocking action, each one linked from the place the user gets stuck.

How do we know our demo is too long? Look at the drop-off curve, not the average watch time. Averages hide the answer. You are looking for a cliff, a single point where 30% of viewers leave together, and that point is almost never the end. It is a specific dull moment you can cut.

Does an explainer follow the same rules? No. Explainers are more consistent, usually 60 to 120 seconds, because they run where the viewer knows the least. The formats differ in more ways than runtime, which we break down in explainer video versus product demo, and in our guide to SaaS video marketing.