Creative Performance Metrics for DTC Video Ads: Hold Rate, Thumbstop, and Thruplay Benchmarks (2026)

By the end of this guide you will know what each metric measures, the formula behind it, the current third-party benchmark ranges (and why you should treat any benchmark with suspicion), and how to use these diagnostics to explain a CPA that looks broken when the media buying is actually fine.

This is the measurement layer that sits under creative strategy. For the creative and targeting playbook that sits on top of it, pair this with our Meta ads strategy guide for DTC brands.

What Are Creative Performance Metrics for Video Ads?

Creative performance metrics are the in-view engagement signals that tell you how a video ad holds attention, separate from the money metrics like CPA and ROAS. CPA tells you the ad failed. Creative metrics tell you why.

Think of a video ad as a funnel inside the ad itself. An impression serves. Some people watch 3 seconds. Fewer make it to 15. Fewer still finish and click. Each drop-off point has a metric attached to it, and the shape of that drop-off is the diagnosis. A great hook with a terrible hold means your first 3 seconds oversell and the body disappoints. A weak hook with a strong hold means the concept works but the opening is skippable.

The reason this matters for DTC specifically: on autoplay feeds, most of your budget is spent on people who never really watch. Attention is the scarcest thing you are buying. Buffer's analysis of 1.1 million TikTok videos, cited by eMarketer, found a median watch time of just 6.9 seconds on 30 to 60 second videos. You are usually fighting for the first few seconds, not the last.

Hook Rate (Thumbstop): The First Three Seconds

Hook rate, or thumbstop rate, is the percentage of impressions that turn into a 3-second video play. It answers one question: did the thumb stop scrolling?

Hook rate = 3-second video plays / impressions

Meta defines a 3-second video play as the number of times your video was played for at least 3 seconds, or for nearly its total length if it is shorter than 3 seconds (Meta Business Help Centre). Divide that by impressions and you get the share of the feed that gave your ad a real look.

This is the metric that governs everything downstream. If the hook fails, nothing else in the ad gets a chance. Motion, which analyzes DTC creative at scale, states that 73% of ecommerce video ads fail within the first three seconds. Whether the exact figure holds for your account or not, the direction is the point: most of the loss happens before the pitch even starts.

Here is where the benchmarks land, and the honest caveat comes first: these are third-party agency benchmarks, not platform-official numbers, and they move with vertical, placement, audience temperature, and objective. Use them as a rough gut check, not a target.

Notice the two sources disagree on where good starts, which is exactly why a single benchmark is dangerous. Prospecting on a cold audience in-feed will hook fewer people than a Reels placement or a warm retargeting pool. Judge a hook against your own account's history and against the other ads in the same test, not against a blog number.

Hold Rate: Do They Stay Past the Hook?

Hold rate measures retention after the hook. Of the people who watched 3 seconds, how many were still there around 15 seconds later? This is where you find out if the body of the ad earns the attention the hook won.

Hold rate = 15-second video views / 3-second video plays

That denominator matters, and it trips people up constantly. Some teams calculate hold rate off impressions instead, which produces a completely different (and much smaller) number. Before you compare your hold rate to anyone else's, confirm you are both dividing by the same thing. A 12% hold rate off impressions and a 12% hold rate off 3-second views are not the same ad.

Using the retention definition, AdSights puts DTC Meta feed prospecting hold rate around 12-25%, with a median near 18%, and Reels a touch higher near 23%.

The pattern to read: a high hook and a low hold is the most common failure mode in DTC video. The opening promises something the ad never delivers, so people bail the second they realize it. The fix is not a new hook. It is a body that pays off the promise faster.

ThruPlay Rate: The Completion Signal

ThruPlay is Meta's completion metric. According to Meta's Business Help Center, a ThruPlay counts when your video plays to completion if it is shorter than 15 seconds, or for at least 15 seconds if it is longer. ThruPlay rate turns that into a percentage.

ThruPlay rate = ThruPlays / video plays

On autoplay placements, video plays roughly track impressions, so in practice this reads as what share of the feed effectively finished the ad. AdSights puts a 15 to 30 second DTC prospecting ThruPlay rate around 15-30%, with a median near 22%.

ThruPlay is also an optimization event, not just a diagnostic. You can tell Meta to optimize delivery toward people likely to watch to completion, which is useful for upper-funnel awareness and view-through work. Just do not confuse a high ThruPlay rate with sales. Completion is an attention signal. Whether that attention converts is a separate question you answer with CPA, ROAS, and incrementality testing, which we cover in our guide to incremental sales lift and omnichannel measurement.

The DTC Video Metrics Cheat Sheet

Here is the whole diagnostic funnel in one place. Treat the benchmark column as directional third-party ranges, not targets.

Benchmark sources: AdSights glossaries, Taylor Sicard's 2026 Meta benchmarks, and the eMarketer/Buffer analysis for watch time. Every one of these is an aggregate across accounts that look nothing like yours, so the value is in the shape of the funnel, not the exact digits.

How Creative Diagnostics Explain a Broken CPA

This is the part media buyers care about most. A rising CPA sends everyone hunting for a bidding or audience problem, when the real culprit is often sitting in the creative funnel.

Run the diagnosis top to bottom. If the hook rate cratered, the creative stopped stopping thumbs, usually fatigue or a weak opening frame. If the hook is fine but the hold collapsed, the body is the problem. If hook and hold are both healthy but CPA is still ugly, now you have permission to look at the landing page, the offer, the audience, or the attribution window rather than blaming the ad.

We have inherited plenty of accounts where a failing campaign had a perfectly good hook rate and a hold rate in the basement. The previous team kept relaunching new hooks and nothing moved, because the leak was 8 seconds in, not at the start. Reading the metrics in order would have saved months.

The discipline is to never judge a video on CPA alone. CPA is a lagging, noisy, multi-cause number. The creative metrics are earlier, cleaner, and they tell you which lever to pull.

Reading Video Metrics on TikTok

TikTok has its own version of the same funnel, and the language differs. Instead of hook rate you will hear about 6-second views and view-through rate (VTR), and completion metrics matter more because the format rewards watch time.

The honest state of TikTok benchmarks: reliable public numbers are thin, and a lot of TikTok benchmark figures floating around are untraceable. What TikTok does publish is optimization uplift rather than absolute rates. TikTok's Ads Help Center, for example, reports that its View+ optimization drove an average 62% uplift in 6-second VTR on standard Feed. That is a lift from a tool, not a benchmark you should hold your account to.

The transferable principle is the same across platforms: watch time is the currency, the first few seconds decide most of the outcome, and completion is an attention proxy, not a sale. For the full channel approach, see our TikTok ads strategy guide for DTC brands.

How Jetfuel Uses Creative Diagnostics

We treat creative metrics as the earliest, cheapest signal we have, so we read them before we touch budgets. Every video in a test gets judged on hook, hold, and completion first, and only then on CPA, because the creative funnel tells us what to fix while CPA just tells us something is wrong.

Our practice is to diagnose in order and act on the specific leak. A dead hook gets a new opening frame or a fresh concept, not a budget cut. A strong hook with a weak hold gets a re-edit of the body, not a new thumbnail. When hook and hold are both healthy and the money metrics still lag, we stop blaming the creative and move the investigation to offer, landing page, audience, and attribution. That sequencing is the difference between iterating with a reason and reshuffling ads and hoping. It also keeps our creative tests honest, because we are comparing ads against the account's own baseline, not against a benchmark pulled from someone else's portfolio.

Frequently Asked Questions About DTC Video Ad Metrics

The Bottom Line

Creative metrics are the earliest read you get on a video ad, and they are far more actionable than CPA on its own. Hook rate tells you if the scroll stopped, hold rate tells you if the body earned the attention, and ThruPlay rate tells you if people effectively finished. Read them in that order and a broken campaign usually points you straight at the real leak.

Just hold every benchmark loosely. The numbers in this guide are third-party ranges across accounts that are not yours, useful for a gut check and useless as a target. Your own baseline is the benchmark that counts.