What follows is a consolidation of the current benchmarks Meta and TikTok have published for paid social creative performance. These are not estimates or industry averages. They come directly from Meta for Business and TikTok for Business documentation. They represent what the platforms themselves define as good, and they are the standards your creative is being measured against every time an ad goes live.
Hook Rate — Meta
Meta defines hook rate as the percentage of people who watch at least three seconds of your video after seeing it in their feed. It is the first signal the algorithm uses to decide whether your ad is worth distributing further.
The benchmark ranges Meta uses: below 20% is underperforming, 20 to 25% is baseline, 20 to 40% is good, and above 40% is excellent. Most ads that brands consider successful are sitting in the 20 to 40% range. The ones that compound reach over time tend to be above 30%.
What this means practically: if your ad does not create a reason to keep watching in the first three seconds, roughly 75% of the people who see it will scroll past before the algorithm has a chance to learn who it should be shown to. The hook is not a creative preference. It is the price of entry.
Hold Rate — Meta
Hold rate, sometimes called ThruPlay rate, measures how many of the people who watched three seconds of your video watched it through to completion. Meta's target range is 40 to 50% or higher for sub-15 second ads.
This metric matters because the algorithm uses it as a quality signal on top of the hook rate. An ad that hooks people but does not hold them suggests the opening lied about what was coming. An ad that hooks and holds tells the algorithm that the content delivered on its promise, which is the profile the algorithm rewards with broader distribution.
Watch Time and Completion — TikTok
TikTok's distribution engine is built around maximizing watch time across the platform. The FYP algorithm treats watch completion as its primary signal for deciding whether content deserves wider distribution.
For paid ads on TikTok, the target completion rate on sub-15 second ads is 70% or higher. For longer formats, even a 50% completion rate is considered strong. The 3-second view rate benchmark is 71% — meaning 71% of viewers make their decision to watch or scroll within the first three seconds of any ad.
The practical implication is that TikTok ads need to be built differently than Meta ads. On TikTok, the opening moment is not just the hook — it is the entire argument for why this content deserves a viewer's attention.
Format and Aspect Ratio
Format alignment is one of the most consistently underestimated performance variables in paid social creative. The data is unambiguous: 9:16 vertical format is required for TikTok FYP and Meta Reels. 4:5 outperforms 1:1 in Meta Feed by up to 15%. 1:1 performs poorly in Reels placements because it signals to the viewer that this content was not made for this format.
The practical implication is that most brands are either producing creative in the right formats and then using it in the wrong placements, or producing for one format and hoping it works everywhere. Neither approach holds up when the data is this clear about what the algorithm rewards.
Sound-Off Viability
A significant portion of social media consumption happens with sound off. TikTok data shows that adding captions to ads produces an estimated 32% completion lift. Meta research suggests that ads which can communicate their core message without audio significantly outperform those that cannot.
Sound-off viability is not about accommodating deaf users, although that matters too. It is about the fact that your ad is being served in environments where sound is not on, and the algorithm is measuring whether people are still watching.
Authenticity Signals
Both platforms have published data showing that creator-led, platform-native content outperforms polished brand production in distribution metrics. On Meta, UGC-style content outperforms high-production creative by up to 40% on reach efficiency. On TikTok, creator-fronted ads show a 193% higher 6-second view rate than brand-produced alternatives.
The important nuance here is that native feel and low production quality are not the same thing. What the platforms are rewarding is content that looks like it belongs in the feed — that does not visually announce itself as an ad before a viewer has decided whether to watch.
What to Do With This
These benchmarks are most useful not as targets to optimize toward, but as a diagnostic framework. When a piece of creative underperforms, these numbers tell you where to look first. Low hook rate means the opening three seconds are not working. Low hold rate means the content promised something it did not deliver. Low completion rate on TikTok means the pacing is losing people somewhere in the middle. Poor format alignment means the creative was built for a different context.
The brands that consistently hit these benchmarks are not necessarily producing better creative. They are producing creative that is built to meet these standards before it is launched, not rebuilt after the data comes back.
That is the fundamental shift that pre-launch creative intelligence is designed to support — moving the diagnostic from after the spend to before it.