-
how many ugc pieces does it take to find a real winner. our hit rate is ~1 in 12 and i don’t know if that’s normal
Posted by on May 26, 2026 at 9:21 pmtracking some of our 2026 numbers and trying to figure out if our creative hit rate is bad, normal, or actually good.
setup: dtc supplement brand, ~$120k/mo paid social spend, mostly meta + some tiktok. we test ugc creative continuously, refreshing about 8-12 pieces per month.
over the last 90 days:
– 34 ugc pieces tested
– 5 became "stickers" (in paid for 3+ weeks, beat baseline cpa)
– 3 became "workhorses" (in paid for 6+ weeks, scaled to >$1k/day each)
– 26 got cut within 2 weeks
so a 1-in-7 sticker rate, 1-in-12 workhorse rate. is that good?
industry numbers i can find are all over the place. some agencies claim 1-in-3 winner rate (which i don't believe). some performance marketers on linkedin say their winner rate is 1-in-20 (which sounds low). the reality has to be somewhere in this range but nobody publishes real data.
what i can tell you about the 8 that worked vs the 26 that didn't:
– the workhorses came from 3 different creators, none of them our most expensive
– 6 of 8 winners had the same structural hook (problem stated within first 2 seconds, product reveal at 3-4 seconds)
– only 1 of the 26 cut creatives followed that structure
– ad copy length didn't matter as much as creative hook
– broad audiences killed weaker creatives faster (which is actually useful, it surfaces winners sooner)
what i'm trying to figure out:
-
is 1-in-7 sticker / 1-in-12 workhorse normal for paid social in 2026?
-
is there a hit rate above which we should be willing to pay premium per piece, vs continue testing high volume cheap?
-
if anyone's tracking a higher winner rate, what's different about their approach?
would love to hear from anyone running $50k+/month paid spend with visibility into their actual numbers, not theoretical "best practices."
-
-
1 Reply
-
1 in 12 for a true workhorse honestly sounds pretty healthy to me, esp in supplements where fatigue hits fast and compliance limits what u can say.
the thing that stands out is not the creator cost, its the hook pattern. if 6 of 8 winners used problem in first 2 sec then reveal at 3 to 4 sec, id treat that as the brief, not a random insight. keep creator pool broad and cheap ish, but pay more for ppl who can reliably hit that structure without sounding scripted.
imo the better metric is not winner rate overall, its winner rate by tested concept. if u made 34 pieces but only like 8 had the strongest hook format, ur actual hit rate on the right format might be way higher than 1 in 12. the 26 losers may be telling u more about briefing quality than creative quality.
-
1 in 12 workhorse doesn’t sound broken to me at that spend, especially if broad is doing the filtering for you. The bigger signal is that 6 of 8 winners shared the same first-2-second hook, so I’d turn that into a tighter test matrix before paying more per creator. I’ve been using KREV AI for this exact ecommerce creative workflow because it helps spin new UGC angles around the winning structure instead of ordering another batch and hoping one lands.
-
Depends how quuickly you can iterate from Champion UGC templates
-
I think thats normal 1 in 12 is the new normal.
-
1-in-12 for actual scalable workhorses honestly sounds pretty normal to me. A lot of people inflate “winner” numbers by counting creatives that only hold for a few days.
The bigger signal is probably the shared hook structure across your winners. That’s the part I’d double down on more than chasing huge creative volume.
-
One in twelve is actually solid for UGC. Most teams see one in twenty or worse. Your issue might not be hit rate but how you’re defining a winner. Are you measuring views, clicks, or actual conversions?
Log in to reply.