If you only watch one metric to catch creative fatigue early, watch hook rate. On Meta ads it moves before cost per acquisition, before frequency alarms, and before your dashboard tells you anything is wrong. It is the closest thing you have to a smoke detector for a fading ad.
The catch is that most teams either ignore hook rate or misread it. They average it across a campaign, confuse it with thumbstop rate, or treat a drop as a reason to panic-refresh when the smarter move is often to pivot. This is a short, practical guide to what hook rate on Meta ads actually is, how to calculate it, what good looks like in 2026, and what to do when it starts to slide.
Key takeaways
- Hook rate is three-second video views divided by impressions. It is the earliest attention signal on a Meta video ad.
- In 2026 the rough consensus is 25 percent baseline, 30 percent good, 35 percent before you scale, and below 20 percent means the opening is broken.
- Hook rate is not thumbstop rate. Thumbstop measures the pause, hook rate measures whether the pause survived into attention.
- It declines before CPA does because attention fades at the top of the funnel while SKAdNetwork lag delays the bottom.
- A drop is a decision point, not a panic button: refresh within the angle if the angle still works, pivot the angle if the zone is exhausted.
What hook rate on Meta ads actually is
Hook rate is the percentage of impressions that turn into a three-second video view. The formula is simple: three-second video plays divided by impressions, times 100. If 1,000 people were served your ad and 280 watched past the three-second mark, your hook rate is 28 percent.
Ads Manager does not give you hook rate as a default column, which is part of why so few teams track it. You build it as a custom metric: take the three-second video plays column, divide it by impressions, and save it as a calculated column you can pull into every view. Read it at the individual ad level. Averaged across an ad set or a campaign it becomes meaningless, because one strong ad will mask three that are quietly dying.
Why three seconds and not one? Because three seconds is roughly the point at which a passive scroll becomes an active choice to keep watching. One second tells you the thumbnail stopped the thumb. Three seconds tells you the opening actually earned attention. That is the thing worth measuring.
What a good hook rate looks like in 2026
Benchmarks vary by source and vertical, but the 2026 consensus across performance teams is fairly tight. Treat around 25 percent as a baseline, 30 percent or higher as good, and 35 percent or higher as the level you want before pouring budget into an ad. The median sits near 28 percent, and anything north of 40 percent puts a creative in the top tier of what advertisers are seeing. Below 25 percent, rework the first frame. Below 20 percent, the hook is underperforming and belongs in a test against alternatives, not in your scale campaign.
Adjust for context. Impulse consumer apps and DTC-style offers should comfortably clear 30 percent, because the feed reflex is doing half the work. Considered-purchase and B2B categories can run sustainably at 20 to 25 percent, as long as the hold rate further down the video is strong. A hook rate in isolation is a number. A hook rate against your own category and your own history is a signal.
Hook rate vs thumbstop rate
These two get used interchangeably and they should not be. Thumbstop rate is typically built on one-second or two-second views: it tells you the first frame earned the pause. Hook rate uses the three-second view: it tells you the pause turned into attention. You can hold thumbstop rate steady while hook rate falls, and that specific pattern is diagnostic. It means your opening frame still earns the half-second of curiosity but the next two seconds are no longer converting it. The audience is starting to recognise the ad and check out before it gets going. That is early fatigue, and thumbstop rate alone would never have shown it to you.
Why hook rate declines before CPA does
Hook rate is a top-of-funnel measurement, so it reacts the instant an audience begins tuning out a creative. Cost per acquisition is a bottom-of-funnel measurement, and for mobile apps it is delayed further by SKAdNetwork postbacks that arrive 24 to 72 hours after the event, sometimes longer. Stack those two facts together and the timeline is unforgiving: attention drops today, installs soften over the next few days, and your CPA dashboard only confirms the problem a week later once the postbacks have caught up.
That week is expensive. Running a fatiguing creative for an extra seven days at scale is one of the most common and avoidable ways to burn budget on a mobile app account. Hook rate is how you buy that week back. When it starts trending down while your CPA still looks fine, believe the hook rate. It is the leading indicator; CPA is the lagging one.
What to do when hook rate drops
A falling hook rate is a decision point, not an instruction to make another ad. The move depends on why it is falling:
- Refresh within the angle. If this is one ad fading while others in the same psychological zone still perform, the angle is fine and the execution is tired. New opening, new creator, same underlying idea. This buys you weeks.
- Pivot the angle. If several ads built on the same angle are all sliding in parallel, the zone is saturated and a fresh visual will fatigue just as fast. You need a genuinely different psychological angle, not another variation of the same one.
- Do not panic-refresh. Swapping the visual on a saturated angle feels productive and changes nothing. It is the single most common mistake we see, and it is why refreshes so often fatigue as quickly as the ad they replaced.
The reason the pivot decision matters more than the refresh decision is that hook rate fatigue is usually a coverage problem, not an execution problem. If your library lives in one or two psychological zones, every ad in it competes for the same worn-out attention. We unpack the full mechanism, and the framework for fixing it, in our guide to why refreshing your Meta ads won't fix creative fatigue and in the case for signal diversity over volume. If you want a quick estimate of how much fresh creative your spend actually needs, the creative refresh calculator is the fastest way to get one.
Frequently asked questions
What is a good hook rate on Meta ads?
The 2026 industry consensus is roughly 25 percent as a baseline, 30 percent or higher as good, and 35 percent or higher as the threshold most performance teams want before they scale spend. Below 20 percent the opening is underperforming and should be tested against alternatives. These are guides, not laws: considered-purchase categories can run sustainably at 20 to 25 percent if hold rate is strong, while impulse consumer apps should be clearing 30.
How do you calculate hook rate in Ads Manager?
Hook rate is three-second video views divided by impressions, expressed as a percentage. Ads Manager does not show it as a default column, so build a custom column dividing the three-second video plays metric by impressions, or export both and calculate it yourself. Read it at the individual ad level, never at the campaign level, because averaging hides the ad that is actually fading.
What is the difference between hook rate and thumbstop rate?
Thumbstop rate is usually based on one-second or two-second views and measures whether the opening frame earned the initial pause. Hook rate uses the three-second mark and measures whether that pause survived into actual attention. Thumbstop tells you the thumbnail worked; hook rate tells you the first few seconds held. Hook rate is the more useful fatigue signal because it declines while thumbstop can still look healthy.
Why does hook rate drop before CPA rises?
Hook rate measures attention at the very top of the funnel, so it reacts the moment an audience starts tuning out a creative. Cost per acquisition sits at the bottom of the funnel and, for mobile apps, is further delayed by SKAdNetwork postbacks of 24 to 72 hours. By the time CPA confirms a problem you have often been running a fading ad for a week. Hook rate gives you that week back.
Does a high hook rate guarantee a profitable ad?
No. Hook rate only measures whether you won attention, not whether you converted it. A clickbait opening can post a strong hook rate and still fail on installs and Day-7 ROAS. Use hook rate as an early warning and a diagnostic, then always read it alongside install rate and downstream value rather than treating it as a standalone target.
Want this run for you?
Tracking hook rate is the easy part. Knowing whether a drop means refresh or pivot, and having the psychological coverage to pivot into, is the work. That is what we build for the mobile apps we run creative for at The Social Outline.
If you run a mobile app spending real money on Meta and want a creative system that reads the early signals instead of the lagging ones, apply to work with us. We take a small number of mobile app clients per quarter.