ThruPlay is Meta's count of times a video was played to completion or for at least 15 seconds, and cost per ThruPlay is amount spent divided by ThruPlays. A 3-second video play needs 3 seconds of play, or 97% of a shorter video. Hook rate is 3-second plays divided by impressions; hold rate is ThruPlays divided by 3-second plays. Meta now defines both ratios itself but labels them in development, while thumbstop rate and view rate remain buyer shorthand. This guide separates Meta's columns from calculated ratios, reads them as a funnel, covers when ThruPlay optimization fits and sets up reporting across many accounts.
Meta's video metrics vs the ratios buyers calculate
In Ads Manager, the Video engagement preset under Columns holds Meta's video metrics, including impressions, ThruPlays, cost per ThruPlay and video plays. Plays are counted per impression. Video plays counts starts, autoplay included, replays excluded. Duration metrics such as 3-second plays and ThruPlays count unique seconds played, so rewatching a segment adds nothing, while milestone metrics such as video plays at 25% also count people who skipped ahead. On a carousel, the video columns sum every video in the ad.
Ratios are where definitions drift. Meta now defines two: hook rate, 3-second plays divided by impressions times 100, and hold rate, ThruPlays divided by 3-second plays. Both are labeled in development; if your column picker lacks them, the same formulas work on the base columns. Thumbstop rate, thumbstop ratio, view rate and video view rate appear in neither Meta's glossary nor its video metric pages, so ask for the numerator and denominator before comparing anyone's number with yours.
- Video plays, and 2-second continuous video plays, which count only after 2 continuous seconds, usually with at least half the video's pixels in view
- 3-second video plays and cost per 3-second video play
- ThruPlays and cost per ThruPlay, plus 6-second ThruPlays
- Video plays at 25%, 50%, 75%, 95% and 100%
- Video average play time, which includes replays
- Hook rate and hold rate, Meta's own ratios, labeled in development
ThruPlay and cost per ThruPlay
Meta defines ThruPlays as the number of times your video was played to completion, or for at least 15 seconds. Its glossary entry, which now also calls it 15-second ThruPlays, adds the mechanics: a shorter video counts once at least 97% has played, the metric measures how much of the video played rather than time elapsed, and replays within an impression don't count. Meta treats 97% as full length because viewers often drop off as credits roll.
Cost per ThruPlay is amount spent divided by ThruPlays, and Meta labels it in development. ThruPlay is also a performance goal at two durations: 15 seconds, or 6 seconds, counted at 6 seconds of play or 97% of a shorter video. The 6-second ThruPlays column is labeled in development too. An ad set chasing 6-second ThruPlays isn't chasing 15-second ones, so check the goal before comparing costs.
The 97% rule makes ThruPlay length-sensitive. A 10-second video earns one at about 9.7 seconds of play, a 60-second video only at 15 seconds. Cost per ThruPlay therefore tends to flatter short edits: fine when attention is the goal, misleading when you rank videos of different lengths. Compare within length bands, and read it as the price of attention, not of customers.
3-second plays, hook rate and thumbstop rate
A 3-second video play counts each impression where the video played for at least 3 seconds, or 97% of a shorter video, again counting content played and ignoring replays. In Meta's examples, a 2-second video played for 1.94 seconds counts, while watching the first 2 seconds of a 3-second video twice does not. Cost per 3-second video play divides amount spent by that count.
Hook rate turns that count into a share of impressions; Meta says it measures how well the opening 3 seconds grab attention. The denominator is broad: Meta counts an impression when the ad first appears on screen and, outside Audience Network, the video doesn't have to start playing, so a video that never starts lowers hook rate as surely as a weak opening. Thumbstop rate is the same idea under an industry name; if your team divides by something else, put that in the column name. 2-second continuous video plays give a different early-attention rate, so never mix 2-second and 3-second rates in one comparison.
There is no good hook rate in the abstract. A blended figure mixes placements that present video differently, and very short videos earn 3-second plays at under 3 seconds. Set a baseline from recent ads in the same account, placement, format and length band, and judge each new opening against it. Our creative testing guide covers test structure and sample size.
Hold rate, percentage watched and average play time
Meta's hold rate divides ThruPlays by 3-second plays: of the plays the opening held for 3 seconds, how many went on to a ThruPlay. Meta describes it as plays to completion, but because a ThruPlay counts at 15 seconds, on longer videos it measures who reached 15 seconds, not the end. Buyers use other versions, such as ThruPlays over impressions, so write the formula into the column name and keep it fixed.
The milestone columns at 25%, 50%, 75%, 95% and 100% show where viewers leave. Divide each by 3-second plays for a retention curve: an early drop points at the beat after the hook, a late one at the offer or ending. Milestones include people who skipped ahead, and percentages scale with length: 25% of a 12-second video is 3 seconds, of a 60-second video 15, so on videos under 12 seconds the first point can top 100%.
Video average play time is total play time, replays included, divided by video plays, so Meta notes it can exceed the video's length. Looping is one way that happens: Meta says Instagram videos loop continuously while on screen, and Facebook videos of 30 seconds or less loop for up to 90 seconds, sometimes 180. Compare it only between similar lengths on the same placement.
Reading the funnel: hook, hold, click, convert
Read the metrics in the order a viewer meets the ad. Hook rate says whether the opening stopped the scroll, hold rate and the milestone curve whether the body kept people, link click-through rate whether the offer moved them, and cost per result whether the click became business. Each stage points at one part of the edit or the page, instead of a vague verdict that the creative isn't working.
Upstream metrics are diagnostics, not goals. A shock opening can lift hook rate while pulling in viewers who never buy, and a short video can post a high hold rate simply because it's short. When a video with a lower hook rate wins on cost per purchase, the purchase decides. Track hook and hold per creative over time too: a hook rate sliding as frequency climbs can be an early sign of fatigue, which our frequency and ad fatigue guide covers.
- Weak hook, everything else normal: test new openings on the same body
- Strong hook, weak hold: the beat after the hook loses people, so tighten or reorder it
- Strong hold, weak click-through: the offer or call to action is unclear or comes too late
- Strong click-through, weak conversion: check the landing page, offer and audience before the video
When ThruPlay optimization makes sense, and when it doesn't
ThruPlay is a performance goal in the Awareness and Engagement objectives, for auction and reservation campaigns. The 15-second goal can bill on impressions or per ThruPlay, the 6-second goal on impressions. Meta describes a performance goal as the outcome its delivery system bids on, so a ThruPlay ad set looks for people likely to watch. Sales, Leads, Traffic and App promotion don't optimize for video views, though they can run video ads.
It fits when attention is what you're buying: awareness flights, story-led launches, video series, or pools of viewers to retarget. It doesn't fit as a stand-in for conversion testing. The people most likely to watch aren't necessarily the people a Sales ad set finds, so a video that wins on cost per ThruPlay can lose on cost per purchase. Test creative where the budget will run, and never compare cost per ThruPlay across different performance goals.
Placement changes completion too. Meta's Ads Guide says Instagram Stories plays videos under 16 seconds in full but may split longer ones into separate cards, showing one, two or three automatically, tailored to each viewer; with a ThruPlay goal, viewers may get a Keep watching option. Facebook Stories splits videos over 15 seconds and offers Play Full. Completion on long Stories videos partly reflects how many cards each viewer saw, so break results out by placement and cut a Stories edit of 15 seconds or less.
Video metrics Meta has removed or changed
Meta's About metrics being removed page records the big video changes. In July 2019 ThruPlay replaced 10-second video views as the default optimization for video views campaigns, and the 10-second view metrics were removed from 29 October 2019. The same update renamed views to plays, such as 3-second video views to 3-second video plays and average watch time to average play time, without changing the calculations.
The video views objective went too, folded into Awareness and Engagement when Meta cut 11 objectives to 6. The page's latest entry, from October 2024, removed unique metrics such as unique purchases, not video columns. Not every change is a removal: Meta has also added the 6-second ThruPlay goal and the 6-second ThruPlays, hook rate and hold rate columns. Audiences changed too: as of October 2026, video engagement custom audiences offer three groups, all viewers at 3 seconds, engaged at 10 seconds and highly engaged at 50%, so audience tiers no longer line up one to one with the ThruPlay and milestone columns.
Watch the in-development label, which also covers cost per ThruPlay, 2-second continuous plays, video plays at 100% and average play time. Meta says these metrics may still be adjusted and recommends them for directional guidance, not historical comparison or important business decisions. For long-run baselines, store counts that don't carry the label, such as impressions, 3-second plays and ThruPlays, and compute your own ratios.
Reporting video metrics across many ads and ad accounts
Build one column preset for video, with impressions, 3-second plays, ThruPlays, the milestones, average play time, link clicks, results, amount spent and ad ID, tick Save as a column preset, and have every buyer save the same set. Keep a one-page sheet with the formula behind each ratio, including whether the team reports Meta's hook rate column or its own calculation.
For cross-account views, open Ads Reporting from your business portfolio and select several ad accounts for one report; you need admin, advertiser or analyst access. Cross-account reports have limits: Meta's examples of unsupported metrics include results and video watches, data goes back 24 months instead of 37, the time zone defaults to the business portfolio's, and they can't be emailed on a schedule. Check that your video columns load before relying on one. Single-account reports can be scheduled by email or exported, and the Ads Insights API returns nearly any metric available in Ads Manager for your own warehouse.
Break results down by placement, video view type or video sound; where available, the Ad creative breakdown compares creatives, excluding dynamic creative ads. Aggregate counts, not ratios: sum impressions, 3-second plays and ThruPlays across ads and accounts, then recompute hook and hold rate from the totals, since averaging ad-level percentages lets small ads swing the result. Don't add up reach or other unique metrics, which Meta deduplicates. Include deleted ads through the Delivery filter, or totals come up short, and export the history you want to keep before it leaves the reporting window.
Following one video across ad sets and accounts is easiest with the same creative ID in every ad name; our naming conventions guide and our creative asset management guide set that up. Agency ad accounts fit the same workflow: Power Ads creates them in its own business portfolio and shares them into your Business Manager through partner access, so once your people are assigned to them they use the same columns and reports. The cost is 4% of each top-up with no setup fee, and creative and its analysis stay with your team.
Key takeaways
- Video plays, 3-second plays, ThruPlays, milestone plays from 25% to 100% and average play time are Meta's core video columns; every ratio is built on them.
- A ThruPlay is a play to completion or for at least 15 seconds, so cost per ThruPlay favors short videos; compare it within length bands.
- Meta now defines hook rate and hold rate itself, labeled in development; thumbstop rate and view rate are buyer terms, so check their formulas.
- No universal benchmark exists: judge hook rate and cost per ThruPlay against your own baseline for the same account, placement, format and length.
- ThruPlay optimization buys attention, not customers, so test creative for conversion campaigns in conversion-optimized ad sets.
- Across accounts, sum the counts, recompute the ratios and carry a creative ID in every ad name; cross-account reports cover 24 months and can't be emailed on a schedule.
FAQ
What is a ThruPlay on Meta?
A ThruPlay is counted when your video is played to completion or for at least 15 seconds; videos shorter than 15 seconds count once 97% has played. Replays within one impression don't add to it. Meta also offers a 6-second ThruPlay goal, and its 6-second column is labeled in development.
How is cost per ThruPlay calculated?
Amount spent divided by ThruPlays. Meta shows it in the Video engagement preset and labels it in development. Because short videos earn ThruPlays at 97% of their length, it tends to look better on short edits, so compare it only between similar lengths, placements and performance goals.
What is hook rate, and is it the same as thumbstop rate?
Hook rate is 3-second video plays divided by impressions, times 100: the share of impressions where the video played for at least 3 seconds. Meta now defines it and labels it in development. Thumbstop rate or ratio is industry shorthand for the same idea; check the denominator before comparing reports.
What is the difference between hook rate and hold rate?
Hook rate measures the opening; hold rate, ThruPlays divided by 3-second plays, measures what happens next. Because a ThruPlay counts at 15 seconds, hold rate on longer videos shows who reached 15 seconds, not who finished; the 75%, 95% and 100% milestone columns show the ending.
What is view rate on Meta ads?
View rate isn't in Meta's glossary or video metric pages, so it means whatever the person reporting it chose, such as 3-second plays or ThruPlays divided by impressions. Name the column after its formula and don't compare view rates across teams until the formulas match.
Do agency ad accounts report video metrics differently?
No. Meta's video metric definitions are the same whichever business owns the ad account. Agency ad accounts are shared into your Business Manager through partner access, so your team reads them with the same columns and reports; if their time zone differs from your other accounts, cross-account reports use the business portfolio's time zone. Power Ads supplies accounts this way and doesn't make or analyze creative.
Sources
- facebook.com/business/help/1792720544284355
- facebook.com/business/help/1868286323447328
- facebook.com/business/help/471190536725647
- facebook.com/business/help/1796060333844808
- facebook.com/business/help/1441620301326177
- facebook.com/business/help/743427195703387
- facebook.com/business/help/1591938568393543
- facebook.com/business/help/1216404939613119
- facebook.com/business/help/metrics-labeling
- facebook.com/business/help/1794520550789165
- facebook.com/business/help/1517440995178744
- facebook.com/business/help/863736740000947
- facebook.com/business/help/898977192761293
- facebook.com/business/help/1592060448672946
- facebook.com/business/help/592054664510127
- facebook.com/business/help/1925542547764376
- facebook.com/business/help/447834205249495
- facebook.com/business/help/1160942077279936
- facebook.com/business/help/2051461368219124
- facebook.com/business/help/355670007911605
- facebook.com/business/help/1695754927158071
- facebook.com/business/help/1635030749917644
- facebook.com/business/ads-guide/update/video/instagram-story
- facebook.com/business/ads-guide/update/video/facebook-story
- facebook.com/business/help/1099865760056389
- facebook.com/business/help/181580096038351
- facebook.com/business/help/1452248935049010
- facebook.com/business/help/924647297736315
- facebook.com/business/help/2285603948358687
- facebook.com/business/help/675548706216201
- facebook.com/business/help/596403700379202
- facebook.com/business/help/264160060861852
- facebook.com/business/help/243916866413404
- facebook.com/business/help/276601619787641
- facebook.com/business/help/1098122253564910
- developers.facebook.com/documentation/ads-commerce/marketing-api/insights
