Skip to content
Book a call
Glossary

Hold rate

Hold rate is the share of people who keep watching a video ad after its opening. A common way to calculate it is ThruPlays divided by 3-second video plays, which shows whether the body of the ad holds attention.

Also called
Hold ratio, Video hold rate
Category
Creative performance
Updated

How hold rate is calculated

Hold rate is built from two video metrics Meta reports. ThruPlays count the times a video was played to completion, or for at least 15 seconds. For a video shorter than 15 seconds, Meta counts a ThruPlay once 97% of it has played. 3-second video plays count the times a video played for at least three seconds, or for 97% of a video shorter than that. Both sit in Ads Manager under the video engagement columns.

Divide ThruPlays by 3-second video plays and you get the share of people who got past the hook and kept watching. Meta measures both by how much of the video played, rather than how much time passed. Rewinding to watch the same seconds again within one impression does not add to either number, so a viewer can't inflate your hold rate by replaying the opening.

Formula

Hold rate = ThruPlays ÷ 3-second video plays

Both inputs come from Meta: ThruPlays (plays to completion or for at least 15 seconds) and 3-second video plays (plays of at least three seconds); hold rate itself is an industry convention that Meta does not list among its video ad metrics, and some teams divide ThruPlays by impressions or use video plays at 50% instead.

Why hold rate matters for brands advertising on Meta

A strong hook gets people in. Hold rate tells you whether the ad deserved their attention once they stayed. The middle of an ad is usually where the product, the proof and the reason to buy appear, so a viewer who leaves at the fourth second never sees the part that sells.

Hold rate also stops you from fixing the wrong thing. When the hook works and the hold rate is weak, the opening probably made a promise the rest of the ad did not keep, or the pace dropped once the hook was over. When both are healthy and sales are weak, the creative has done its job, and the offer or the landing page needs a closer look. If you only look at sales, you can't tell these cases apart. Each answer leads to a different fix, and a different person on your team doing it.

Other ways teams measure hold

Meta does not list hold rate among its video ad metrics, so teams measure it in different ways. Some divide ThruPlays by impressions, which mixes the hook and the body into one number. Some use Meta's milestone metrics, such as video plays at 25%, 50% or 75% of the video's length. Meta notes that these milestones include people who skipped ahead to that point, so they show how far into the video people got, which can differ from how long they watched. Others use video average play time, which Meta calculates with replays included.

Hold rate only makes sense for video, since a static ad has no watch time. Within video, each choice answers a slightly different question. What matters is picking one definition and using it the same way across every ad you compare. A hold rate from one formula can't be compared with a hold rate from another.

What a good hold rate looks like

We don't give a target number, because hold rate depends heavily on the length of the video. A ThruPlay asks for 15 seconds whatever the length, unless the video is shorter than that. So a 10-second ad earns a ThruPlay only when it is almost finished, while a 60-second ad earns one at the 15-second mark. Two ads of very different lengths are running different races. The format, the audience and the offer change the picture too.

Compare like with like instead. Put ads of a similar length, shown to similar audiences, side by side, and look at where each one loses people. The milestones at 25%, 50% and 75% can show the stretch where viewers leave, and that tells an editor which part to tighten or cut.

How Fifth Dimension uses hold rate

We read hold rate next to hook rate in the review after each launch. Together they show whether an ad lost people at the door or once they were inside.

A weak hold usually sends us back to the ad script. We check whether the body pays off the hook, whether the product appears early enough and whether any line drags. Then we look at the edit, because slow pacing loses people even when the words are right. Who is speaking matters too. For Herbar, we paired creator footage with voiceover from the co-founder. When one ad holds people much better than the rest, we study what happens in its middle and carry that into the next brief. Our Creative Strategy work covers that whole loop, from research and scripts to production and the review of results that shapes the next round.

Sources

  1. Meta Business Help Center: 15-second ThruPlays
  2. Meta Business Help Center: 3-second video plays
  3. Meta Business Help Center: About video ad metrics on Facebook and Instagram
  4. Meta Business Help Center: About video ad metrics calculation

Questions about Hold rate

What is the difference between hook rate and hold rate?

Hook rate measures how many impressions turn into three seconds of viewing, so it judges the opening. Hold rate measures how many of those viewers go on to a ThruPlay, so it judges the rest of the ad. You need both to see where an ad loses people.

What counts as a good hold rate?

It depends on the length of the video, the format, the audience and the offer, so we don't use a fixed target. Compare ads of a similar length shown to similar audiences, and study the ones that keep people watching longest.

How do you improve a low hold rate?

Start with the script. Make sure the body delivers what the hook promised, bring the product in sooner and cut any line that repeats itself. Then tighten the edit where the milestones show people leaving.

Put these ideas to work on your ads

Book a call and we'll look at your product, your current ads and what to test next.