What is hook rate in video advertising?

Short answer. Hook rate is the share of a video ad's impressions in which the viewer kept watching past the opening, most often calculated as 3-second video plays divided by impressions, times 100. It is a practitioner metric, not a platform one, so definitions vary and you should state yours.

There is no single official definition

None of the ad platform help pages we checked defines a metric called hook rate. It is built by advertisers and analytics tools from the video metrics platforms do report. Two published versions:

• Motion, a creative analytics company, says thumbstop rate and hook rate "are the same metric" and gives the formula "(3-Second Video Plays ÷ Impressions) × 100 = Hook Rate %".
• Growth consultant Marcus Burke, interviewed by AppsFlyer in November 2025, uses "hook rate (3s video views / impression)".

The numerator is the same idea in both, but the platforms count views differently. TikTok, for example, reports "2-second video views", defined as the "Number of times your video starts playing for at least 2 seconds in an impression session", with replays excluded. A TikTok hook rate built on 2-second views is not comparable to a Meta one built on 3-second plays.

Hold rate disagrees too

The companion metric, hold rate, shows the definitions drifting further. Burke uses "hold rate (thruplays / impressions)". Motion uses "(Through plays ÷ 3-Second Video Plays) × 100 = Hold rate". Same name, different denominator.

A worked example (illustration)

These numbers are made up. A video gets 100,000 impressions, 25,000 3-second plays and 5,000 ThruPlays.

• Hook rate: 25,000 / 100,000 = 25 percent.
• Hold rate by Burke's definition: 5,000 / 100,000 = 5 percent.
• Hold rate by Motion's definition: 5,000 / 25,000 = 20 percent.

The same ad has a 5 percent or a 20 percent hold rate depending on whose formula your dashboard uses. Write the formula next to the number.

What a good or bad reading tells you

A low hook rate says the opening is not stopping people: the first frame, the first line or the sound. A high hook rate with poor results further down says the opening promised something the rest of the ad, or the page after it, did not deliver. We would not publish a benchmark here: the published ones are vendor opinions that vary by platform, audience and format.

In games and in consumer brands

The metric works the same way in both, but it only exists where the ad is a feed video with a skip-by-scrolling viewer. In our view, on full-screen in-app placements, where the viewer cannot scroll past, a 3-second view is close to automatic, so hook rate tells you much less there.

Common mistakes

• Comparing hook rates across platforms with different view thresholds.
• Optimizing for hook rate, which rewards shock openings that do not sell.
• Changing the opening and the body at once, so you cannot tell which part moved.

Flixr's view

We use hook rate as a diagnostic, never as the decision. Hook rate, hold rate and thumbstop are diagnostics, not outcomes, which is the argument of measure creative on purchases, not on the ad. A hook is a promise; our taxonomy of hooks covers how to write ones that keep it.