Intro
Your ad is stopping the scroll. Meta's telling you as much — hook rate's healthy, thumb-stop ratio looks fine on paper.
Conversions are still flat.
Most people's next move is "test more creative." That's the expensive way to learn something your own retention curve would've told you for free.
Worth having open while you read this
Your ad's actual retention graph, inside Ads Manager (Ads Reporting → breakdown by video play %). Most people check hook rate and thumb-stop ratio and stop there. The shape of the drop-off between second 3 and the end is where the real diagnosis lives.
A quick word from us
Reading a retention curve properly is a five-minute skill that saves weeks of blind creative testing. If you want a second pair of eyes on your account's actual numbers before your next brief, that's exactly the kind of work we do. [See how we work with DTC creative teams →]
The mechanism
Treat every ad as having two separate psychological jobs, not one. Job one: Stop — interrupt the scroll, earn the first three seconds. Job two: Convince — build enough trust, desire, or urgency in what follows to actually move someone to buy.
Hook rate only measures job one. It tells you whether the opening frame or line worked. It says nothing about job two.
So a high hook rate with flat conversions isn't a creative failure. It's a specific failure — the opening did its job, and something after it didn't. Treating that as "the creative doesn't work" and starting from scratch throws away the one part that's already proven itself.
How to actually locate the failure
Pull up the retention curve — the percentage of viewers still watching at each second of the video — and look at its shape, not just the endpoint.
A steep cliff right after the hook. Most people bail in the first few seconds after the hook lands. This usually means the opening made a promise or opened a curiosity gap the body didn't pay off fast enough. It felt like bait. Fix: tighten the transition from hook to payoff, don't rewrite the hook itself.
Retention holds steady, drop only near the very end. The creative is doing its job the whole way through. If conversions are still flat here, the problem almost certainly isn't the ad — it's the offer, the landing page, or the price. Fix: look downstream of the ad before touching the ad again.
A slow, steady bleed with no clear cliff. Nobody's leaving because of one bad moment — they're just losing interest gradually. Usually a pacing problem in the middle section: too much setup, not enough forward motion. Fix: cut the middle, not the ends.
Three different shapes, three different fixes, and none of them is "make new creative and hope."
What this looks like in practice
Imagine two ads for the same supplement brand, both with a strong 55% hook rate. Ad A's retention falls off a cliff at second 4, right after the hook — the promise made in the first line never gets addressed. Ad B holds retention at 40% all the way to the CTA, but conversions are still low. Same top-line metric, completely different problems: Ad A needs a rewritten body, Ad B needs a look at the actual product page it's sending people to.
The actual decision to make this week
Before you brief a single new piece of creative, pull the retention curve on your best-hooking, worst-converting ad and find where it drops. That single graph tells you whether you're about to fix the right problem — or spend a week's ad budget rewriting the one part of the ad that was never broken.
Before you go
Got a mechanism you want decoded — a metric that feels off, a claim from a guru that sounded too clean? Reply to this email. Future issues get built from what you're actually stuck on, not what's trending.