The folklore is settled: user-generated content outperforms polished brand video, so shoot on a phone and stop overthinking it. It's repeated so often that it's stopped being a claim and started being an assumption.
The largest piece of evidence on the question says something more awkward.
That is the opposite of what most creative strategy decks assert. It is worth sitting with before explaining it away.
Why both things can be true
The resolution isn't that UGC is bad. It's that "UGC beats brand video" is a claim with no funnel stage attached, and the answer changes completely depending on which stage you mean.
Work summarised by the California Management Review puts it plainly, citing Colicev et al. (2019): UGC does its work during the brand awareness stage, and much less so for persuasion. It earns attention and credibility from people who don't yet know you. It is not reliably the thing that closes.
Which makes the two findings compatible:
- Top of funnel — a creator holding your product in their kitchen reads as a recommendation rather than an ad. That's where UGC's credibility advantage lives.
- Lower funnel — someone who already knows you and is deciding whether to buy needs information quality, clarity and proof. That is exactly where the meta-analysis found firm-generated content stronger, and it is what a well-made brand asset is built to deliver.
The meta-analysis measures brand loyalty and its antecedents from survey respondents, not incremental revenue from paid social. It also finds the effects are moderated by product involvement, hedonic versus utilitarian category, market development and platform — which is a formal way of saying "it depends on your category". Do not read it as "make less UGC". Read it as "the blanket claim doesn't survive contact with a large sample".
The question that's actually useful
"UGC or brand-made" is the wrong axis, because it bundles together several things that vary independently: who is on screen, how polished the production is, whether the account owns the asset, and what the ad is trying to do.
Split them and the analysis gets far more useful. A founder talking to camera and a paid creator talking to camera share a format but differ in source. A studio product demo and a handheld product demo share an angle but differ in production values. Those distinctions are where the actual learning is, and collapsing them into one binary throws it away.
| Instead of asking | Ask |
|---|---|
| Does UGC beat brand video? | At which funnel stage does each win in this account? |
| Should we make more UGC? | What share of our spend sits behind owned assets versus partnership assets, and how does each perform against the metric its campaign was built for? |
| Is this creator working? | Is it the creator, the angle, or the hook that's working — and which of those transfers to the next ad? |
The owned-versus-partnership question
For performance brands running both, there's a second-order issue the UGC debate obscures entirely: dependency. If the ads carrying your spend are overwhelmingly partnership assets, your performance is renting someone else's credibility, and your cost base moves when their rates do. If they're overwhelmingly owned, you may be under-using the one format that earns cheap attention at the top.
Neither is wrong. But you can only manage the balance if you can see it, and most reporting can't distinguish the two at all.
Knack derives owned versus partnership from the ad itself — the Instagram account the creative was published from — so the split is there before anyone tags anything, and it doesn't depend on a naming convention. It also grades each creative against the metric its campaign was actually built for, so an awareness-objective UGC ad is never marked down for lacking ROAS it was never meant to produce. That is the mechanism that keeps the funnel-stage distinction in this article from collapsing back into a single leaderboard.
What to do on Monday
- Split your creative by funnel stage first, source second. Judge each group on its own metric.
- Check what share of spend sits behind partnership assets. If it's most of it, that's a strategic exposure, not a creative one.
- Stop testing "UGC vs brand". Test the variables underneath it — talent, setting, pacing, angle — because those are the things a brief can actually specify.
Sources
- Tyrväinen, O., Karjaluoto, H., & Ukpabi, D. (2023). Understanding the role of social media content in brand loyalty: A meta-analysis of user-generated content versus firm-generated content. Journal of Interactive Marketing, 58(4), 400–413. 223 independent samples, 97,709 respondents.
- California Management Review (November 2024). Is User-Generated Content (UGC) a double-edged sword for marketers? — summarising Colicev et al. (2019) on UGC's effect being concentrated at the awareness stage.