What Is Hold Rate? Definition and How to Read It
Hold rate is the share of viewers still watching a video ad at 15 seconds, ThruPlays divided by impressions. Learn how to read it alongside hook rate.
Updated June 2026 · Likit Sae Lee, CTO
Hold rate is a video ad metric that measures the share of viewers still watching at the 15-second mark, calculated as ThruPlays divided by impressions. Where hook rate (3-second video plays divided by impressions) shows whether your opening stopped the scroll, hold rate shows whether the body of the video kept people watching after the hook. It is not a default Ads Manager column, so advertisers typically build it as a custom metric.
What hold rate measures
Hold rate answers a simple question: of everyone your video ad reached, how many were still watching at the 15-second mark? It is calculated as ThruPlays divided by impressions, expressed as a percentage. The definition leans on how Meta counts a ThruPlay. According to Meta's documentation, a ThruPlay is recorded when a video is played to completion if it is 15 seconds or shorter, or when someone watches at least 15 seconds of a longer video. Meta counts unique seconds watched, so time spent replaying the same segment is excluded, and it treats at least 97% of a video's length as a full view because people routinely drop off during end cards. Dividing ThruPlays by impressions turns that raw count into a rate you can compare across ads with very different reach.
Hold rate vs hook rate
Creative teams almost always read hold rate next to hook rate. Hook rate is 3-second video plays divided by impressions, and it measures the first moment: did the opening frame and first seconds stop the scroll? Hold rate measures what happens after that moment: did the video earn continued attention once someone paused on it? The two together diagnose where a video loses people. A high hook rate with a low hold rate suggests the opening made a promise the body did not keep, or the pacing sagged right after the hook. A low hook rate means the problem sits in the first frames, and hold rate barely matters yet because few viewers ever reached second four. Neither term is an official Meta metric name; both are conventions built from metrics Meta does report.
How to see it in Ads Manager
Hold rate is not a default column in Ads Manager. To track it, create a custom metric that divides ThruPlays by impressions, then add it to your reporting columns, or export your video metrics and calculate it in a spreadsheet. For a deeper look at where viewers leave, pair it with the video play percentage metrics Meta reports at milestones such as 25%, 50%, 75%, and 95%, which show the shape of drop-off rather than a single checkpoint. AdPlay.ai's creative analytics tracks this kind of retention comparison across your ads, which makes it easier to line up several creatives side by side without rebuilding custom metrics for every ad account.
When hold rate matters, and when it does not
Hold rate matters most for videos meaningfully longer than 15 seconds, where the 15-second checkpoint sits partway through a story: UGC testimonials, demos, and narrative ads live or die on whether viewers hold past the hook. For videos of 15 seconds or shorter, a ThruPlay requires completion, so hold rate effectively becomes a completion rate and reads much stricter than the same figure on a long video. Because of that, only compare hold rates between ads of similar length, format, and placement, and against your own account history rather than someone else's numbers. Watched over time on a single ad, a sliding hold rate is one of the earlier signals of creative fatigue, often showing up before cost metrics visibly worsen.
Frequently asked questions
Is hold rate an official Meta metric?
No. Meta reports the two ingredients, ThruPlays and impressions, as standard metrics, but it does not report hold rate as a named column. The term comes from creative strategists and media buyers who divide one by the other to get a comparable percentage. In Ads Manager you can build it yourself as a custom metric using ThruPlays divided by impressions, or export your data and calculate it in a spreadsheet. Because it is a convention rather than an official definition, always check how any report or tool defines it before comparing numbers.
What is the difference between hold rate and hook rate?
Both are ratios against impressions, but they measure different moments. Hook rate uses 3-second video plays, so it tells you whether the first frames stopped the scroll. Hold rate uses ThruPlays, which Meta counts at completion for videos of 15 seconds or shorter and at 15 seconds watched for longer videos, so it tells you whether people stayed once they stopped. A strong hook rate with a weak hold rate usually means the opening over-promised and the body under-delivered. A weak hook rate means most people never reached the part hold rate measures.
What is a good hold rate?
There is no universal figure, and any single benchmark hides too much. Hold rate shifts with video length, placement, audience temperature, and format, so a 60-second narrative ad and a 15-second offer ad are not comparable on the same scale. The reliable way to use it is relative: compare new creatives against your own account history for the same format and placement, and watch the trend on a single ad over time. A hold rate that declines on a previously stable ad is often an early sign of creative fatigue.
Keep exploring
Turn ad research into winning ads
Research the ads that work, generate the creative on-brand, and launch to Meta, all in one tool.
7-day free trial · No credit card required