Which metrics do brands actually judge UGC by?

"What results have you gotten?" — and your stomach drops, because you've delivered twenty videos and never seen a single number on any of them. Here's the answer worth memorizing: brands judge UGC ads by four metrics — hook rate, hold rate, CTR, and CPA or ROAS — and only the first two are verdicts on you.

  • Hook rate — the percentage of people still watching after three seconds. A grade on your first line and first frame. Yours.
  • Hold rate — how many are still there around the fifteen-second mark (a "ThruPlay," in Meta terms). A grade on your pacing and structure. Also yours.
  • CTR — click-through rate. Shared custody: your call to action matters, but so do the offer and the audience the ad was shown to.
  • CPA and ROAS — cost per acquisition and return on ad spend. These belong to the price, the landing page, and the media buyer. You influence them from a distance; you don't control them.

So when a brand says a video "didn't perform," ask which number. A clip with a 35% hook rate and a weak ROAS didn't fail as creative — something after the click did.

What counts as a good hook rate or hold rate?

Rough shape most media buyers would recognize: a hook rate around 20–25% is par, 30%+ is strong, and 40%+ is the clip that gets scaled. Hold rate measured against impressions: under 8% is soft, 8–15% is solid, and above that you've earned a screenshot.

One caveat before you quote any of this: definitions wobble between teams. Hook rate is usually 3-second plays divided by impressions, but some buyers measure hold rate against 3-second plays instead, which flatters it — ask how they define it before comparing.

The bigger truth: the account average is the benchmark that actually matters. A 26% hook rate in an account averaging 19% is a win; the same 26% where the account sits at 33% is a miss. Niche, product, and placement move these numbers so much that global benchmarks are only good for a sanity check.

Hook rate is also the metric you can move on purpose — stronger hooks are the highest-return edit in UGC.

How do you get performance data when the brand runs the ads?

Here's the awkward truth of this job: you make the ad, but the brand owns the dashboard — so by default, you never see your own results. The fix isn't a hack, it's a habit: plant the ask at handoff, then follow up two to three weeks after the ads go live, asking for two numbers instead of dashboard access.

At delivery, one line does the planting: "Would love to hear how these run — even ballpark numbers help me make your next round stronger."

Then the follow-up, once the ads have real spend — the wait matters; day-two numbers are noise:

Hi Maya — hope the launch went well! Quick one: now that the ads have some spend behind them, could you share the hook rate and hold rate on my clips, and roughly how they compare to your account average? Ballpark is totally fine — it helps me make your next round better.

Every clause has a job. "Next round" frames the data as serving the brand, not your ego; two numbers keeps the reply under a minute — a media buyer can pull them before their coffee cools; "ballpark" removes the excuse to postpone. About half of brands answer, in my experience, and the silence from the rest is workload — not a hidden bad review.

How do you turn one result into a mini case study?

Three lines: context, number against a baseline, consequence. That's the entire format —

Skincare brand, TikTok ads, 30-second product demo. 32% hook rate vs. their 21% account average. They scaled the ad and booked two more rounds.

The context proves it's real client work. The baseline turns a floating stat into a verdict — 32% alone is trivia; eleven points above the account average is proof. And the consequence (scaled, rebooked) shows the brand voted with its budget. Only got one number back? Lean harder on the consequence line.

Then put it where it gets read: under the video it describes in your portfolio, mid-pitch in your outreach emails, on your media kit. One honest case study sitting next to the work it's about does more than a page of adjectives.

The quiet payoff: a creator with a documented result is no longer priced like a gamble — and that's the cleanest opening you'll get for raising your rates.

What do you say when you have no numbers yet?

A brand asks for results and you have nothing to show — every working creator has stood exactly there. Say the true thing, then redirect: "Most of my clients run the ads on their side, so I don't always see final numbers. Here's what I can show you." Then show proof you actually own:

  • Organic proxies. Retention graph, average watch time, completion rate from your own TikTok or Instagram analytics — real evidence you can hold attention, clearly labeled as organic so nobody thinks it's ad data.
  • Process numbers. Hook variations tested per concept, turnaround time, revision count — "most videos approved with zero or one revision."
  • Rebooks. "Three of my last four clients came back for a second round" is a performance metric. Brands don't rebook creative that flopped.

One hard line: never invent a number. The person reading your pitch stares at dashboards all day; a made-up ROAS reads fake in seconds, and it burns exactly the trust the real numbers were supposed to build.

And remember the long game. Every creator whose page now says "32% hook rate" started with zero data and one slightly awkward follow-up email. Send yours after the next delivery — the numbers stack up faster than you'd think.

FAQ

Should I ask for access to the brand's ad dashboard?

No — ask for two numbers instead. Dashboard access means permissions friction most brands won't bother with, and hook rate plus hold rate against the account average tells you nearly everything about your creative.

What if the brand never shares numbers?

Nudge once, then let it go — plenty of teams are simply busy, and silence isn't a review. Keep building proof you control and make the ask again with the next client; the habit only has to land half the time to fill a page with results.

Do organic views count as UGC results?

Yes, as labeled proxies. "This hook held 70% of viewers past three seconds organically" is honest and useful. Present organic wins as organic — implying they were paid results crosses into inventing stats.

Can I share a brand's metrics publicly?

Ask first. A one-line permission request usually gets a yes when you stick to creative metrics like hook rate rather than revenue figures. If you can't get a yes, anonymize — "a skincare brand" with real numbers still works as proof.