A couple of years ago, I launched a redesign of the ad experience for a platform with over 100 million monthly active users. We showed users more ads, more frequently. And the complaints about ads went down. Counterintuitive, isn’t it? Let me explain.
In the UX industry, we often treat “friction” and “annoyance” as the same thing. We shouldn’t. Let’s differentiate them.
Friction is structural. It’s any step, delay, or interruption that prevents a user from reaching their goal. By default, it’s a neutral experience. Whether a loading screen or a confirmation dialog annoys the user depends entirely on context – when it appears, whether the user expected it, and how much control they have over it.
Annoyance is the emotion users feel when friction appears at the wrong moment in the user journey.
You can have friction without annoying users. You can have extremely annoyed users without any designed friction in place.
Still, “reducing friction” is often a UX mantra that designers follow. It sounds logical at first, but you do need a healthy amount of friction. Either to benefit the user (e.g., confirmation screens before making a bank transfer) or to make the business stay afloat (e.g., by getting revenue from converting users or showing ads).
Whether friction causes annoyance or not depends heavily on the timing and context, and ultimately, on the perception users have.
The original ad experience on our platform worked like this: a user would land on a question page, read the content, hit a blocker, click a CTA to unlock more, and then watch a 30-second video ad.
Annoyance came from:
The new ad strategy flipped the experience completely.
We hit them with the ad as soon as they landed on the page. The page didn’t even fully load yet, and we already threw an ad at our users.
“How annoying!” you might think. Well, it wasn’t. Let’s unpack why:
Before, an average visitor was exposed to ~30s of ad time for every 20 answers they saw. With the new experience, they were exposed to roughly ~50s of ad time per 20 answers they saw.
Almost twice as much ad time. “Twice as much friction”. Yet the engagement and retention increased.
The reason our new model worked is that even though people watched many more ads, it FELT like they watched fewer of them.
Showing a 30-second ad after an explicit CTA made users expect an ad and then spend the whole 30 seconds intentionally watching the ad.
Showing 10x 5s ad on page load hits different. Before users oriented themselves, 3 seconds had already passed. Then, seeing the “2s left” and navigating the mouse over it to be ready to click it just felt like an automatic action for many.
Not to mention, most users already expect a ~1s delay when launching a new page for content to load. By loading the ad first, we could grab that 1 second for free.
In the end, five seconds of ad time felt like a second for users. Thanks to that, we could show users many more ads over time.
The most surprising thing, though, and I still can’t wrap my head around why it happened, is that our %CTRs on ads dropped by only 30% (we expected like an 80% drop!). My theory is that people decide if they are going to click on the ad or not in the first second, and then the rest of the ad doesn’t really matter.
The hardest part of the ad redesign wasn’t the design work. It was convincing the AdOps and sales teams.
Their 30-second fullscreen video was the platform’s most premium placement with the highest CPM and the most demand. Their bonuses were tied to premium placement revenue. Our metrics were tied to engagement and retention. Same product, different objectives.
My first instinct was to argue the UX case. This went nowhere.
What moved the conversation was using their own language. Instead of “this is a better experience”, I tied how a potential boost in retention and engagement would eventually break even (or maybe even improve) their ad revenue metrics.
I built projections showing that shorter, more frequent ads across more pageviews would generate more total revenue in the long run, thanks to engagement boost (less perceived friction = people visit us even more often = we show them even more ads). The math wasn’t guaranteed, but it was credible.
That was a bet that short-term loss in ad revenue will recover in the long-run.
Although it wasn’t enough to convince the sales team, it was enough to convince the CEO to try out the new strategy 🙂
Let’s recap. We showed more ads, more often, without any warning, right after users landed on our website. By all the “best case practices” and UX principles, it’s a recipe for failure. It’s “a lot of repeated friction before giving any value without any warning”.
Yet, it turned out to be the biggest user experience improvement we made so far.
Right context, right timing, and right math turned the most annoying part of our user experience into a mildly irritating automated action that users do.
The takeaways?
But you don’t innovate by following the crowd.
LogRocket's Galileo AI watches sessions and understands user feedback for you, automating the most time-intensive parts of your job and giving you more time to focus on great design.
See how design choices, interactions, and issues affect your users — get a demo of LogRocket today.
Through this step-by-step guide, you can learn how to easily import files from Adobe Illustrator to Figma.
Growth design isn’t just about getting more users to convert. It’s about using research, experimentation, and data to improve activation, retention, and long-term user value without sacrificing trust or the quality of the experience.
Starting with proto-personas can be better than a blank page, but don’t forget — they’re assumption-driven placeholders for the real thing. Research is key to turning them into true personas.
Design thinking workshops are your key to turning big problems into clear solutions. In this blog, I share how to run them efficiently and keep your team aligned.
Learn more →
