Ugly Ads Beat Pretty Ones | Scrolling Media
Why lo-fi, ad creative outperforms polished production in app advertising. Data, examples, and a framework for making ugly ads that actually convert.
March 20, 2026
You spent $3,000 on a creator shoot. Color graded it twice. Added motion graphics your editor spent a full day on. The ad looks gorgeous.
It also tanked.
Meanwhile, some competitor filmed a vertical video on their iPhone in a parking lot, slapped some native TikTok text on it, and their CPA is half of yours. If this sounds familiar, congratulations: you’ve discovered the ugly ad problem.
Here’s what nobody in your creative team wants to hear: Meta’s algorithm in 2026 is heavily weighted toward engagement signals, not visual quality. When an ad looks like an ad, users scroll past it before the algorithm gets a chance to learn anything useful. No early engagement means the system reduces distribution confidence. Your beautiful creative dies in the auction before it even competes.
Raw, unpolished content generates those early engagement signals faster than polished brand creative. The algorithm sees likes, comments, shares, and saves happening quickly, so it distributes the ad more aggressively. That’s the entire game right now.
Motion’s analysis of $100 million in ad spend found that 42% of top performing DTC ads deliberately use lo-fi content. Not accidentally. Deliberately. These brands chose to look worse on purpose, and the performance data backed them up.
A study referenced by Rachel Pedersen found that “ugly” Facebook ads achieved conversion rates up to 2.5x higher than polished alternatives targeting the same audience with identical offers. Same audience. Same offer. The only variable was production quality, and the scrappy version won by a landslide.
Barry Hott has built an entire brand around this concept. The guy literally wears a hat that says “Make Ugly Ads.” His approach isn’t about making bad creative. It’s about making creative that doesn’t trigger the part of your brain that says “this is an advertisement, ignore it.”
An ugly ad uses simple text overlays, shaky phone footage, voice overs with slight background noise, and nothing that screams “a brand paid for this.” It blends into TikTok feeds, Instagram Reels, and YouTube Shorts because it looks like something your friend posted, not something a marketing department approved after six rounds of revisions.
The distinction matters: ugly doesn’t mean lazy. It means intentionally choosing authenticity over polish. A piece from Segwise put it well: the best ugly ads focus on pulling people into the problem, showing value fast, and making it easy to take action. No amount of color grading can replace those fundamentals.
If you’re running ads for subscription apps or mobile games, the stakes are higher. AppsFlyer’s Creative Optimization Report analyzed 1.1 million creative variations across $2.4 billion in ad spend and found something wild: tutorials and app reviews generate 45% higher IPM (installs per thousand impressions) and 17% better day 7 retention compared to testimonials. Yet testimonials still capture the majority of budgets.
That disconnect is everywhere in app advertising. Teams keep funding yesterday’s playbook because it’s familiar and easy to produce. But as RevenueCat’s analysis points out, the old UGC formula (ring light, scripted lines, forced enthusiasm) has become so predictable that users scroll right past it. Cost per trial skyrockets, trial to paid conversions crash, and suddenly your unit economics are underwater.
The apps winning right now are the ones running lo-fi creator content that looks like it belongs on your For You Page. Skits, street interviews, pattern interrupt videos, expert commentary shot on an iPhone. SEM Nexus reported that when users scroll through Reels, Shorts, or TikTok, their brains are hardwired to instantly swipe past anything that looks like a commercial. Your ad creative needs to look natively built for the platform. Period.
Here’s where it gets really interesting for performance marketers. Ugly ads aren’t just cheaper to produce (brands typically save 40 to 60% on production costs when replacing studio shoots with creator content). They’re also dramatically faster to test.
When a single ad costs $3,000 and three weeks to produce, you test maybe four concepts per month. When a creator can film five variations in an afternoon for $500 total, you test twenty. The math isn’t subtle.
Dara Denney demonstrated this perfectly with her “Post-it Note Ad” experiment, showing how the most stripped down creative formats can outperform anything over-produced:
View this post on Instagram
Speed matters because creative fatigue is real and accelerating. Shuttlerock’s research found that when you match platform native fonts, text overlays, and shooting styles to whatever feed you’re targeting, engagement rises because the content feels like it belongs there. Not because it looks expensive.
Making ugly ads that perform isn’t an excuse to be sloppy. There’s a framework here.
Start with the hook, not the production. Your first 2 seconds need to stop the scroll. That’s a copywriting problem, not a cinematography problem. Test five different hooks with the same lo-fi footage before you ever think about production upgrades.
Match the platform. TikTok native text overlays on a TikTok ad. Instagram’s aesthetic sensibility on Reels. YouTube Shorts’ pacing for Shorts. Each feed has its own visual language, and ads that speak it perform better than ads that fight it.
Let creators be themselves. The worst thing you can do is hand a creator a word for word script and ask them to perform it like a teleprompter robot. Give them the key message points and the value prop, then let them deliver it in their own voice. The slight imperfections, the natural pauses, the real personality: that’s the whole point.
Test at volume. If you’re only testing two or three polished concepts per cycle, you’re leaving performance on the table. Ugly ads let you run ten concepts in the same time frame for less money, which means you find winners faster and scale them sooner.
Know when polish still wins. Not every ad should be lo-fi. Brand campaigns, retargeting sequences for high value users, and certain premium app categories still benefit from quality production. The point isn’t that polish is dead. The point is that polish should be a deliberate choice based on data, not the default for everything.
This isn’t a trend. It’s a correction. For years, the ad industry assumed that higher production value meant higher performance. That assumption was always shaky, and now the data is killing it.
The brands and apps winning on Meta, TikTok, and Instagram right now share one thing: they stopped optimizing for how ads look and started optimizing for how ads perform. Sometimes those two things overlap. Often, they don’t.
Your creative team might resist this. Ugly ads feel wrong when your job is making things look good. But the numbers don’t lie, and in performance marketing, the numbers are the only opinion that matters.
So next time your $200 iPhone video outperforms your $5,000 production shoot, don’t be surprised. Be grateful you found out before your competitors did.