Best Video Length for Facebook Ads
The best video length for Facebook ads is 15 to 30 seconds for most direct-response campaigns, and under 15 seconds for Reels and Stories. Longer videos are not penalised, but they have to earn every extra second.
Length is the easiest thing to argue about and the least important thing to get right. What decides a video ad is whether anyone is still watching at second three.
What length works for each placement?
Placement dictates format and attention span more than objective does.
| Placement | Aspect ratio | Length that works | Notes |
|---|---|---|---|
| Reels | 9:16 | 6-15 seconds | Fastest scroll, hardest hook |
| Stories | 9:16 | 5-15 seconds | Tap-away is one thumb move |
| Feed | 4:5 or 1:1 | 15-30 seconds | The workhorse placement |
| In-stream | 16:9 | 15-30 seconds | Viewer is already watching video |
| Explainer or demo | 4:5 | 45-90 seconds | Only if the product needs it |
If you run a single video across all placements, build it at 4:5 and keep it under 30 seconds. That is the safest compromise. Better results come from cutting a separate vertical version for Reels and Stories rather than letting Meta crop the feed asset.
The full ad size and spec list covers the technical requirements alongside these.
Why does the first three seconds decide everything?
Because that is where the audience is lost.
The steepest drop in any video ad's retention curve happens immediately. Whatever share of people are still watching at three seconds is roughly the pool available for everything that follows, and no amount of good material at second twenty recovers someone who left at second two.
This is what hook rate measures, and it is the number to optimise before touching length.
Practical consequences:
- Lead with the payoff, not the setup. No logo card, no slow pan, no "hi guys".
- Show the product doing the thing in the first frame if you can.
- Put the text on screen immediately. Most people watch without sound.
- Make the first frame make sense as a still, because that is what a scrolling thumb sees.
When are longer videos the right call?
When the product is unfamiliar, expensive, or hard to believe.
Three cases where 60 to 90 seconds beats 20:
- The product needs demonstrating. A tool whose value only becomes obvious when you see it used.
- The objection takes time to answer. High-ticket services where trust is the barrier.
- Testimonial or founder story creative, where the credibility comes from the person talking, and cutting it to fifteen seconds removes the thing that was working.
A long video that holds attention is a strong signal. Meta's delivery responds to engagement, so a 90-second ad people actually watch can be cheaper to deliver than a 15-second one nobody finishes.
The test is honest: watch the retention curve. If it falls off a cliff at twelve seconds, the ad is a twelve-second ad regardless of how long the file is.
How should objective change the length?
Awareness tolerates length. Conversion rarely does.
- Sales and leads: 15 to 30 seconds. Say the offer, show the proof, ask for the click.
- Traffic: short. The click is the point, so get to it.
- Video views and awareness: longer is fine, because you are buying attention rather than action.
- Retargeting: shortest of all. This audience already knows you, so a 6 to 10 second reminder outperforms a re-explanation.
Retargeting is the length mistake most accounts make. Serving the same 30-second cold-audience explainer to someone who already visited the product page wastes the one thing that audience is short on - patience.
What should you measure instead of guessing?
Three numbers in Ads Manager answer the length question directly.
Hook rate - 3-second views divided by impressions. Tells you whether the opening works.
Hold rate - ThruPlays or 15-second views divided by 3-second views. Tells you whether the middle earns its time.
Average watch time. Compare it against the file length. If a 45-second ad averages eleven seconds, you have an eleven-second ad with thirty-four seconds of waste.
Custom columns make this readable at a glance - see custom columns in Ads Manager for the setup.
Then cut to where the retention curve flattens, not to a number from an article.
What are the common length mistakes?
Four patterns show up constantly in account audits.
A branded intro card. Two seconds of logo animation spends the entire hook window saying nothing. Delete it.
One asset for every placement. A 16:9 video letterboxed into a Reels slot wastes half the screen and signals "advert" before a word is read.
The offer only at the end. If the discount appears at second twenty-eight, most of the people who watched the ad never saw it.
Re-cutting instead of re-shooting. When retention collapses in the first three seconds, the problem is the opening idea, not the edit. A new hook filmed properly beats ten trims of the same footage.
There is a fifth, subtler one: testing lengths and hooks in the same experiment. Change one at a time, or you will not know which variable moved the number. Running the same footage at 15, 30 and 60 seconds is a clean test. Running three different edits at three different lengths is not a test at all.
How do you see what lengths competitors are running?
Watch their ads, and watch which ones survive.
Every active Meta ad is public in the Ad Library, including video, and every card shows the date it started running. An ad running for months is one the advertiser keeps paying for. Collect twenty of those in your category and the pattern in length, pacing and opening frame becomes obvious quickly.
Our free Meta Ad Library downloader saves those videos in original HD with a searchable index, which makes it practical to line up a competitor's whole video library and compare openings side by side. Existing Facebook video ad examples are a faster route if you want the patterns without doing the collection yourself.
Study the structure and the hook. Do not reuse the footage commercially - and this is not legal advice.
FAQ
How long can a Facebook video ad be?
Meta accepts video ads up to 241 minutes in the feed, but that limit has nothing to do with what performs. Feed ads in the 15-30 second range and Reels under 15 seconds cover almost every practical case.
Do shorter Facebook video ads perform better?
Shorter ads usually hold a larger share of their audience, which helps delivery. They do not automatically sell more. A short ad that fails to explain the product converts worse than a longer one that does.
What is the best length for Instagram Reels ads?
Six to fifteen seconds, shot vertically at 9:16. Reels viewers scroll fast, and a cropped horizontal video reads as an ad immediately, which costs you the hook.
Does Meta penalise long video ads?
No. Delivery responds to engagement, not duration. A long video that holds attention can deliver more cheaply than a short one people ignore - the risk with length is losing viewers, not being penalised for it.
Should I make different lengths for different placements?
Yes, where you can. A vertical cut under fifteen seconds for Reels and Stories plus a 4:5 version around twenty-five seconds for the feed covers most of the value. Letting Meta auto-crop a single asset is the weaker option.
Where should the offer appear in a video ad?
Early, and again at the end. Assume most viewers never reach the final frame, so a discount or call to action that only appears at second twenty-eight is invisible to the majority of the people who saw the ad.
The Klipio extension adds a download button to every ad in the Meta Ad Library: one click per ad, or bulk-save a whole search as a ZIP with a searchable swipe file inside. Free, no sign-up.
Get the free extensionKlipio reads a competitor's live Meta ads, ranks them by how long they have been running — the honest signal that an ad is profitable — and turns the winning angle into on-brand creative for your own brand. 3-day free trial.
Start free — 3-day trial →
