Video ad length is the wrong variable to optimise. A 45-second ad that holds attention beats a 12-second one that doesn't, and no rule about duration will fix a weak opening.
That said, there are sensible ranges.
| Placement | Range that works |
|---|---|
| Reels / Shorts, cold audience | 15–25 seconds |
| Feed, cold audience | 15–30 seconds |
| Retargeting | 30–60 seconds — they already know you |
| YouTube in-stream | Say it in five, earn the rest |
| Demonstration or explainer | As long as it stays interesting |
The number to actually watch
Hold rate at three seconds and at fifteen. Together they tell you which problem you have:
- Low at 3s — the hook fails. Nothing after it matters.
- High at 3s, collapse by 15s — the hook wrote a cheque the ad couldn't cash.
- Both healthy, no conversions — the offer or the landing page is the problem, not the video.
Cutting a boring ad in half gives you a shorter boring ad.
Practical rules for video ad length
- Front-load. The thing that matters goes first, not after an intro.
- Captions burned in — assume mute.
- Vertical, always, for social.
- End abruptly. A tidy conclusion signals a script.
- Test the same body with four different openings before you test length. Hooks are the variable that matters.
A worked example
A Gurugram furniture brand ran the same product film in three cuts: 12 seconds, 24 seconds and 48 seconds. The 12-second cut had the best three-second hold rate and the worst cost per enquiry. The 48-second cut had the lowest hold rate at three seconds and the lowest cost per enquiry of the three.
The explanation is not "longer is better". The long cut showed a carpenter measuring a real flat, and anyone who watched past twenty seconds was someone genuinely planning a renovation. The short cut was efficient at collecting attention from people who were never going to buy. Length was acting as a qualifier, which is a job we normally hand to targeting.
That is the useful way to think about duration: every extra second filters the audience a little further. If your product needs explaining or your buyer needs convincing, filtering is good. If the product is obvious and cheap, it is waste.
How to cut the same footage three ways
You rarely need three shoots. You need one shoot and a disciplined edit.
- The proof cut (15s). Result first, product second, nothing else. Use it on cold audiences where you are buying attention from strangers.
- The demonstration cut (30s). One problem, one mechanism, one result. This is usually the workhorse.
- The full story (60s+). Customer speaking, process shown, objection answered. Reserve it for people who already visited the site.
Cut all three from the same session and you have a month of testing from one afternoon of filming.
What we check before blaming the length
- Is the first frame doing work? Most weak ads open on a logo or an establishing shot. Both are dead air.
- Do the captions match the audio? Auto-captions on Indian accents are still unreliable, and a mistranscribed price is worse than no caption.
- Is the aspect ratio right for the placement? A 16:9 film letterboxed into Reels loses a third of the screen and most of the hold rate.
- Is the sound designed for mute? If the punchline is spoken and nothing on screen carries it, most viewers never receive it.
How many new concepts a month your spend actually needs is arithmetic — the creative volume calculator works it out.
Length behaves differently again on TikTok, where the first second decides everything: TikTok ads.
Questions people ask
What is the best video ad length on Meta?
There is no single best length. Short ads hold a higher share of viewers; longer ads carry more argument. In practice 15 to 30 seconds suits prospecting and 45 to 90 suits retargeting, where the viewer already knows who you are.
How long should a Reel ad be?
Under 15 seconds for a cold audience. Reels is a fast surface and the first second carries most of the decision, so a longer cut usually loses more viewers than the extra information wins back.
Does a longer ad cost more to run?
Not directly. You pay for impressions, not seconds. A longer ad costs more when it holds fewer people, because the delivery system sees weaker engagement and the auction prices you accordingly.
What should I watch instead of video ad length?
Hold rate at three seconds and the retention curve. If viewers drop at the same point every time, that moment is the problem — cutting the ad shorter just moves the same weakness closer to the start.