How long should a product video be?

The short answer: 30 to 60 seconds for the video at the top of your homepage or in a launch post. Under 15 seconds for an ad in an Instagram-style feed. A few minutes for a demo that someone asked to see.

Those numbers aren’t arbitrary, and they aren’t fashion. Each one comes from who is watching, what they’re willing to give you, and what the video has to get across in that time. The rest of this post is the reasoning, because once you see it, you can work out the right length for any video yourself.

Count the claims, not the seconds

The mistake most teams make is picking a length first. “Let’s do a 60-second video” becomes a brief, and then the brief gets filled.

It works better the other way round. A product video makes a small number of claims, and each claim needs a certain amount of screen time. Add them up and you have the length.

A typical product video makes about six:

1. The promise, usually your homepage headline.

2. Two to four things the product does, each shown on screen.

3. One piece of proof: a number, a customer count, a logo row.

4. The call to action.

Each claim needs time for two things: reading the line and seeing the product do what the line says. Reading a six-word line at a comfortable pace takes about two seconds. [4] Seeing a piece of interface do something takes another three or four. Call it five to seven seconds a claim. Six claims, plus a moment for your logo at each end, lands between 30 and 45 seconds.

That’s what we see in practice. Our six example films each carry six lines, and they run between 32 and 37 seconds. Nobody set out to make them that long. That’s how long six claims take.

Why not shorter

Six seconds is enough to say one thing, and bumper ads prove that one thing can be worth saying. But a bumper works because it’s reminding someone of something they already half-know.

A stranger on your homepage doesn’t know anything yet. They need to see the product doing something before they’ll believe the headline. Below about 30 seconds you’re usually forced to choose between stating the promise and showing it being kept, and a homepage video that only states the promise is a second headline.

Why not longer

Because nobody is waiting for it.

Nielsen Norman Group’s analysis of how long people stay on web pages found that visitors often leave within 10 to 20 seconds, and that a page has to get its value proposition across in about ten. [5] Your homepage video isn’t playing to an audience that sat down to watch. It’s playing to people who are deciding, fast, whether to keep reading.

The top of the page is where you have the most attention you’ll ever get: in Nielsen Norman Group’s eyetracking study, 57% of page-viewing time went to the area above the fold. [6] But that attention comes in seconds, not minutes.

Video engagement data points the same way. When Wistia analysed more than half a million videos in 2016, engagement held fairly steady up to about two minutes and then dropped sharply. [7] In its 2025 data, videos under a minute were watched to 52% of their length on average. [7] Those are people who chose to press play. A homepage visitor didn’t even do that.

There’s also the loop. A homepage video repeats. A 35-second loop means a visitor who stays for 40 seconds has seen the whole argument once. A two-minute loop means most visitors see the first fifth of it, and the part they miss is usually the proof and the call to action, which come last.

When longer is right

Longer is right when the viewer asked for it. Someone who clicked “Watch the demo”, booked a call or opened your help centre has decided to spend time on you, and a three-minute walkthrough respects that. Wistia’s own 2025 guidance, the same source that favours short videos, recommends one to five minutes for explainers. [7]

The mistake is putting the three-minute video where the thirty-second one should be. The top of the homepage is for strangers. (Screen recording vs. product video goes into why the two formats do different jobs.)

One edit, several lengths

You don’t need a separate production for each placement. Make the 30 to 60 second film first, because it’s the one that has to hold the whole argument. Then cut it down:

• A 15-second version for feeds: the promise, the single strongest product shot, the call to action.

• A 6-second version for bumpers and retargeting: the promise and your logo.

Same lines, same shots, same rhythm, fewer of them. If the long version was built from strong claims, the short versions are just the strongest claims.

The rule

Don’t choose a length. Choose the claims, give each one enough time to be read and seen, and let the length fall out. For the video at the top of a software company’s homepage, it almost always falls between 30 and 60 seconds.

Notes

1. Meta Business Help Center, “Best practices for Instagram video ads”: shorter videos, under 15 seconds, tend to perform best in feed; Stories ads perform best under 10 seconds.

2. Google Ads Help, “About video ad formats”: bumper ads last between 5 and 6 seconds.

3. LinkedIn Help, video ads advertising specifications: videos can run from 3 seconds to 30 minutes, with 15 to 30 seconds recommended so an ad qualifies for all placements.

4. The BBC’s subtitle guidelines recommend 160 to 180 words per minute, about a third of a second per word. BBC Subtitle Guidelines

5. Jakob Nielsen, “How Long Do Users Stay on Web Pages?” (Nielsen Norman Group, 2011), analysing Microsoft Research data on page dwell times.

6. Therese Fessenden, “Scrolling and Attention” (Nielsen Norman Group, 2018).

7. Wistia, “How Long Should Your Next Video Be?” (2016, archived: 564,710 videos, 1.3 billion plays) and “Optimal video length” (2025).