BlogUGC Hook Rate and Hold Rate: What's Good and What's Yours
Creators

UGC Hook Rate and Hold Rate: What's Good and What's Yours.

A 20% to 25% hook rate is typical on Meta in 2026. Learn both formulas, why hold rate can read 5% or 22%, and how to get your own numbers from a brand.

November 12, 2026

Your video may have a row in a spreadsheet you've never seen. Some people who run ads from many UGC creators track them that way, and one described the sheet on Reddit: a row per video, with columns for the creator, hook rate, hold rate, cost per purchase and spend, updated every week. Hook rate tells them how many people your opening stopped. Hold rate tells them how many of those people stayed.

That same commenter explained why those two columns are the fairer way to compare creators: a creator with a worse return on ad spend "might just be getting served to colder audiences," while hook rate and hold rate show whether the creator "grabs attention and keeps it." Both numbers still shift with who sees the ad, but they track the part you make more closely than sales do. That works in your favor. And the opening is where UGC is strong: in DataAlly's 2026 benchmarks for Meta ads, UGC videos had a median hook rate of 21.5%, against 15.0% for studio and motion-graphics video.

Both numbers are also getting easier for brands to see. Meta's Business Help Center now lists hook rate and hold rate as metrics of its own. At the start of November 2026, a trade site, The PPC Times, reported ready-made columns for both showing up in Ads Manager, Meta's tool for running ads, though they may not have reached every account yet. So it's worth knowing how they're worked out, and which part of each one is yours.

What hook rate and hold rate measure

Hook rate is the share of an ad's impressions in which the video played for at least 3 seconds: 3-second video plays divided by impressions, times 100. Hold rate is the share of those 3-second plays that went on to a ThruPlay, which Meta counts at 15 seconds, or at the end of a video shorter than that. The first grades your opening. The second grades what comes straight after it.

One detail makes hook rate a tougher grade than it looks. An impression counts every time the ad is on screen, and Meta says a video doesn't have to start playing for the impression to count. Everyone who scrolled straight past still counts as an impression, which is why a hook rate near 20% is normal, not weak.

Here's the math on a made-up ad: a 30-second video, and the row a brand would see for it after a test.

In the brand's reportCount (made up)
Impressions20,000
3-second video plays4,400
ThruPlays1,000

Hook rate: 4,400 ÷ 20,000 = 22%. Hold rate: 1,000 ÷ 4,400 = 22.7%. On a 30-second video a ThruPlay happens at second 15, so this hold rate covers seconds 3 to 15 and says nothing about the second half of the video.

Until recently, brands built both numbers by hand as custom columns; Meta's own video advertising guide showed how to build the hook rate one in June 2026. The new Help Center entries use the formulas above and are marked "in development," Meta's label for metrics it's still testing and may adjust.

On TikTok: 2-second and 6-second views

TikTok's Ads Manager has no column called hook rate or hold rate. It reports 2-second video views and 6-second video views, and brands build the rates from those. Billo, for one, publishes 2-second views ÷ impressions for the hook and 6-second views ÷ 2-second views for the hold. One quirk: TikTok's 6-second count also includes any view where the person engaged within the first 6 seconds, so it isn't pure watch time.

Because the windows differ (2 and 6 seconds instead of 3 and 15), a TikTok hook rate and a Meta hook rate aren't the same measurement, even under the same name.

Same name, different math

Hook rate is mostly agreed on: 3-second plays divided by impressions. Hold rate isn't. Meta's column divides ThruPlays by 3-second plays, which is also the usual way to work it out. Some teams divide ThruPlays by impressions instead, which gives a much smaller number.

Biddyco, an agency that published its 2026 numbers both ways, shows the gap on the same set of ads. Its median "hold rate" is 4.6%, worked out on impressions. Worked out Meta's way, the same ads kept about 22% of their 3-second viewers to second 15. The made-up ad does the same thing: 1,000 ÷ 20,000 = 5%, against 22.7%.

For the same ad, you can convert one into the other. To turn the impressions version into Meta's, divide it by the hook rate: 5% ÷ 22% = 22.7%. That works because the impressions version is just the hook rate times Meta's hold rate (22% × 22.7% ≈ 5%). A hold rate in single digits is a clue that it was worked out on impressions, so ask before you compare it with anything.

Three more mismatches to watch for:

  • 2 seconds instead of 3. Meta also has a 2-second continuous play metric, so a "hook rate" from a tool built on it won't match one built on 3-second plays.
  • Thumbstop ratio. Another name for much the same idea. Motion, a creative analytics tool, defines it as the share of people watching "the first 2-3 seconds of your video" and notes that it "varies by platform."
  • Your own posts. Your post stats aren't the brand's ad report. Even when a brand runs your video from your handle as a partnership ad, Instagram says your insights on it are the same ones you get for organic posts, while the brand views the ad's performance in Ads Manager. On TikTok, the views, likes and comments a Spark Ad gets are added to your post's own numbers, so the post's counts mix paid and organic. Keep organic and paid numbers apart, in your head and in your pitch.

What's a good UGC hook rate in 2026?

On Meta in 2026, a typical video ad turns between a fifth and a quarter of its impressions into 3-second plays, and about one in five of those viewers stays to second 15. For UGC alone, DataAlly's median hook rate is 21.5%. Each benchmark is one company's read of the ads in its own data, so none of them is the final word. These three state their method (all of them on Meta; TikTok counts other windows), and only DataAlly separates UGC from other video:

SourceWhat it coversHook rateHold rate (Meta's formula)
Billo, first half of 202688,329 Meta video ads run for sales, not split by type, January to June 2026, 14 industries25.44% average; by industry, from 21.32% (pet supplies) to 28.79% (toys and games)Not published
Biddyco479 video ads, not split by type, from the top 50 ads by spend in each of 17 US ad accounts, June to August 202620.9% median; the middle half between 14% and 30%; the top tenth at 45% or more22% median
DataAlly14,219 Meta ads from 45 brands, November 2025 to October 2026; the UGC figures come from the 11 brands that tag their UGCUGC 21.5% median; studio and motion graphics 15.0%UGC 22.8%; studio and motion graphics 23.4%

Billo's figure is an average, which high scorers can pull up; the other two are medians. The made-up ad's 22% would sit right in the middle. Biddyco says it plainly: "If your hook rate is around 20%, you're normal, not failing." In its data, only about a quarter of ads reached 30%.

Two things change what your number means more than any blog benchmark does.

Your category. A pet ad at 23% beats Billo's pet-supplies average (21.32%, the lowest of the 14 industries) while sitting under the 25.44% cross-industry average. Billo's paid marketing lead said the low pet figure was "most likely because the ads never get an animal on screen in the first three seconds." If you film pets, that fix is yours: get the animal on screen in the first three seconds, ideally in the first frame.

The brand's own account. Biddyco found hook rates varying "from a median of about 13% in one account to 41% in another," and its advice is to compare ads "to your own account's median first." That's the fairest bar for your video too, so the brand's usual hook rate is the number worth asking for.

Which part of the number is yours

Hook rate and hold rate grade different seconds of your video, and after second 15 neither one grades anything:

SecondsThe number that grades itWhat moves it
0 to 3Hook rateThe first frame, the first line, motion, on-screen text
3 to 15 (or the end of a shorter cut)Hold rateThe lines right after the hook, and whether they pay off what the opening promised
After 15NeitherThe rest of the video, plus the offer, the price, the landing page and who saw it

Meta's guidance for the first row is specific: "Use motion or a compelling visual in the first frame," and add captions "so viewers can follow the video ad on mute." The levers that make a stranger stop are a craft of their own, and our guide to video hooks covers them.

The second row is where UGC's head start runs out. DataAlly found UGC and studio video about level on hold rate (22.8% against 23.4%), even though UGC wins the opening by a wide margin. So in DataAlly's data, the UGC look helps most with the stop, and the lines right after your hook have to earn the next 12 seconds on their own.

A creator posted exactly this problem in November 2026: "my average hook rate is 38% but my average watch time is 6 seconds." If it's worked out the usual way, that hook rate beats every median in the table. Someone who reviews creator ads replied that the pattern "usually means the hook promised something the next line didn't pay off," and added: "when i review creator ads i look at second 3, not second 1." The reviewer's fix was to show the product working by second 3 and cut every line that isn't proof. A script that plans the beats after the hook is where you catch that gap before you film.

Read together, the two numbers point to different fixes:

Your two numbersWhat it usually points toWhat you can offer
Hook high, hold lowThe opening promised something the next lines didn't deliverA paid recut that puts the payoff straight after the opening
Hook low, hold highSeconds 3 to 15 keep the people who stop; the opening doesn't stop enough of themNew openings on the same body, priced as extra deliverables
Both high, sales lowWhat comes after second 15 (when the product shows up, the call to action), an opening with little to do with the product, or the offer, the price, the landing pageA paid recut of the weak part; if it's the offer, the price or the page, the fix is on the brand's side, not a reshoot
Both lowOften the concept, the brief or the audienceA talk about the brief before any new footage

A high hook rate isn't a sale

It's tempting to read a high hook rate as proof that the video sells. The data doesn't back that up. In Biddyco's median account, the quarter of ads with the highest hook rates got about 1.5 times the click-through rate of the lowest quarter but cost about 1.4 times as much to show, and "They didn't have a lower cost per purchase." Billo puts it in one line: a strong opening "earns you the chance to sell, and the rest of the video still has to do the selling."

That cuts both ways for you. An ordinary hook rate on an ad that sells is a good result, and you can say so. Biddyco's advice to advertisers backs you up: "Don't kill an ad just because its hook rate is average. If it's converting, it's working." And an opening that stops the scroll but has nothing to do with the product can lift the hook rate, and the cost of showing the ad, without lifting sales. That's why Biddyco also warns: "Don't chase hook rate with clickbait."

What isn't yours

Several things move these numbers without touching your video:

  • Who sees it. Both numbers measure what the people shown the ad did, and a warm audience that already knows the brand "can stop for it more readily than a cold one," as one agency puts it. So a hook rate from retargeting (ads aimed at people who already know the brand) isn't comparable with one from new audiences; DataAlly's benchmark, for one, leaves retargeting campaigns out.
  • Where it runs. Placement alone shifts hook rate: an ad shown mostly in Reels and Stories usually scores differently from one shown mostly in lower-cost placements.
  • How early the brand looked. Meta says an ad set (a group of ads the brand runs together) stays in a learning phase until Meta is showing it steadily, which "usually occurs after about 50 results in the week after the ad set's last significant edit." A result is whatever the brand asked Meta to optimize for, usually purchases or sign-ups. Until then, in Meta's words, "your results aren't necessarily indicative of future performance." One creator's client, running ads for the first time, tested two videos for four days and blamed the hook because the click-through rate was low. On a brand-new ad account, four days is an early read, and click-through rate isn't a hook number.

How to get your numbers from a brand

An ad's results live in the brand's Ads Manager. Meta lists "Access to insights on ad performance in Ads Manager" among the benefits of partnership ads for advertisers, and the insights on your side are the ones you'd get for an organic post. So unless your deal says the brand will share results, you'll have to ask. A media buyer with six years on Meta and TikTok ads wrote that it "just surprises me that content creators don't ask or don't know they can ask (in their contracts) for figures on how much traffic or sales the ad/content did."

Ask at three moments:

  1. Before you film, ask whether they'll tell you how the ad does. It belongs with the other questions to ask a brand before filming.
  2. When you deliver, ask to check back later. One creator's habit, once the work is approved, is to open the message with "I'd love to check in in X amount of time to see how this content did." In the creator's words, "it leaves the door open for future partnerships."
  3. After the test. Meta's learning phase usually ends after about 50 results in the week after an ad set is created or last changed in a big way, so follow up no sooner than a week or so after the ad starts.

Start with the small ask. The same media buyer said "Most brands just don't have time to send ad numbers," even when a video does well, so silence doesn't mean your video flopped. They called asking "did this beat your average?" a smart move from a creator, because "Just knowing your video was in the top 10% is huge proof for their next deal!" Two questions a busy brand can answer in one line: did it beat your average, and which version won?

If they're happy to share, ask for the ad's row.

Ask forWhy it helps
Impressions, 3-second plays and ThruPlays, as countsYou can work out both hold rate versions yourself
The dates it ranEarly numbers move around while Meta is still learning
Placements (Feed, Reels, Stories)Placement changes hook rate on its own
New audience or retargetingWarm audiences can stop more readily, so the two aren't comparable
Which cut and opening ran, and how long it isSo you know which version won; on a cut under 15 seconds a ThruPlay means watching to the end, so its hold rate isn't comparable with a longer cut's
The account's usual hook rateIt's the fairest comparison for your number
On TikTok: 2-second and 6-second views, and Video InsightsTikTok's Video Insights breaks clicks and views down by the second and ranks videos against others in the same industry

The row is easier to pull if your name is in the ad's name. One advertiser running UGC from about 15 creators names each ad like creator_product_v1, and a media buyer in the same Reddit thread pointed out that Ads Manager can filter by ad name, so the prefix pulls up one creator's ads. Suggest it when you deliver.

If nobody answers, you can still look the ad up in Meta's Ad Library to see whether it's running. An ad that's still running isn't proof it won, but it's a fair reason to ask again.

If a bonus rides on the numbers

If a brand offers a bonus tied to how the ad performs, hook rate and hold rate are fairer to you than sales, because they sit closer to the part you made. Audience, placement and timing still move them, so write down exactly how the number will be checked:

  • The formula, spelled out in counts (3-second plays ÷ impressions) rather than "the hook rate column," since Meta still labels that metric in development and says such metrics shouldn't drive important business decisions.
  • The dates, the placements, and a minimum number of impressions before the bonus can count (Billo's benchmark, for one, only counts ads with more than 1,000 impressions).
  • The target itself, set against the brand's own account median rather than a figure from a blog. In Biddyco's data, a 30% target would have paid out on about one ad in four.
  • Who reports it, and how: an export or a screenshot of the ad's row.

Keep the bonus on top of a fixed fee for the video, one that's owed whatever the ad does.

Putting the numbers in a pitch

The basics of a results line belong in your media kit. For an ad, add the fields that make the number checkable:

  • the platform and the formula (or "Meta's hook rate column");
  • the dates and the placements;
  • what it's compared with, ideally the brand's usual hook rate;
  • which cut or opening it was;
  • whether it's one ad, or your typical result across several.

A sample line, with made-up numbers: "Hook rate 31% on Meta (3-second plays ÷ impressions), Reels and Feed, two-week test in March, against the brand's usual 22%. The second of three openings I delivered."

Ask the brand before you share its numbers. If it doesn't want them shared, ask whether relative terms are OK: "beat the brand's usual hook rate by 9 points" names no brand and no spend. Use screenshots only with the brand's OK, and show the clip next to the number only if the brand is happy to be recognized, so the next brand can see the opening that earned it.

Hook rate and hold rate: eight short answers

What is a hook in a UGC video?

It's the opening line and image, the part meant to stop someone scrolling. There's no fixed length: Meta's video ad guide puts the hook at "the first three to five seconds," and TikTok tells advertisers to prioritize it "in the first 6 seconds." Meta's hook rate only measures the first 3, so it grades the very start of your hook.

How do you calculate hook rate?

Divide 3-second video plays by impressions and multiply by 100: 4,400 plays from 20,000 impressions is a 22% hook rate. TikTok reports 2-second views rather than 3-second plays, so a TikTok hook rate is built from those.

What is a good hook rate?

Anything above about 21.5% beats the median UGC ad on Meta in DataAlly's 2026 data, and 30% or more puts an ad in roughly the top quarter of Biddyco's ads. For Meta video ads in general, around 20% to 25% is typical (Biddyco's median for all video was 20.9%, Billo's average 25.44%). The better bar is the brand's own account median and the category you film in.

Is a 35% hook rate good?

On Meta, against these benchmarks, yes. It's above the middle half of Biddyco's ads (14% to 30%), below its top tenth (45% and up), and above every industry average in Billo's data, where the highest was 28.79%. Check the hold rate and the sales too: a 35% opening that loses people at second 4 isn't a winning ad yet.

What does hold rate mean?

It's the share of 3-second plays that went on to a ThruPlay: 15 seconds, or the end of a shorter video. Meta calculates it as ThruPlays divided by 3-second video plays.

What is a good hold rate?

On Meta's formula, two 2026 medians sit at about 22% to 23% (Biddyco's 22% for all video, DataAlly's 22.8% for UGC), so anything above that beats both medians. If someone quotes a hold rate in single digits, ask whether it was divided by impressions: Biddyco's median on that basis was 4.6%.

Is hook rate the same as thumbstop ratio?

Mostly. Both describe the share of people who watch the first seconds. Motion defines thumbstop ratio over "the first 2-3 seconds," while Meta's hook rate uses exactly 3. Ask which one a number is before you compare it.

Does a high hook rate mean the ad will sell?

No. In Biddyco's median account, the quarter of ads with the highest hook rates had a higher click-through rate but also cost more to show, and they didn't have a lower cost per purchase. Hook rate tells you the opening worked; the rest of the video, the offer and the audience decide the sale.

For Creators

Stop the scroll for brands that pay up front

No followers needed, just photos and videos you're proud of. Set your own rates and let brands come to you.

Create your free profile

More to read

All articles