We improved mobile UX on our classified platform

We Improved Mobile UX on Our Classified Platform – Here is Exactly What Worked (and What Did Not)
Last year our mobile bounce rate on Listly hit 68%.
Users would land on a listing, scroll for three seconds, and bounce.
We were losing real money. Mobile traffic had grown to 74% of our sessions, but our conversion rate from mobile was stuck at a pathetic 1.8%. People wanted to buy and sell locally, cars, furniture, services, collectibles, yet the experience on phones felt like trying to shop in a crowded flea market with a blindfold on.
That was the problem.

The Moment We Finally Admitted We Had a Mobile Disaster
I remember pulling up the Hotjar recordings at 2 a.m. and watching dozens of users tap the same “Post Ad” button three or four times before giving up. The button was too small. The photo upload flow took 18 seconds to load on 3G. Our search filters were hidden behind a tiny hamburger menu that people simply never found.
We had launched Listly three years earlier as a bootstrapped classifieds platform, web-first, like most indie projects. Mobile was an afterthought. We slapped on a responsive theme and called it a day. Big mistake.
As traffic shifted to phones, the cracks turned into canyons. Support tickets about mobile UX went from 12% to 41% of our volume in six months. Retention for first-time mobile users was under 19%. We were hemorrhaging potential transactions every single day.

How We Diagnosed the Real Pain Points (The Ugly Truth)
We did not guess. We went straight to the data.
• Session recordings + heatmaps showed 63% of users never scrolled past the first three listings.
• Our in-app survey (simple three-question pop-up) revealed the top complaints: “Cannot easily filter on phone,” “Photos load too slow,” “Hard to message sellers,” and “Posting an ad feels like filling out tax forms.”
• Google Analytics + Mixpanel told us average session duration on mobile was 48 seconds. On desktop it was 3 minutes 12 seconds. The gap was embarrassing.
We also ran five user interviews with real sellers and buyers (we paid them $30 each via UserTesting). Hearing them verbally struggle with our app was painful, but exactly what we needed.

The Changes We Actually Tested (No Fancy Frameworks, Just Brutal Prioritisation)
We could not rebuild everything. We are a small team. So, we focused on the 20% of changes that would move the needle the most.
Here is exactly what we did, step by step:

1. Mobile-First Redesign of Core Flows We rebuilt the listing feed with larger images (optimised to WebP, lazy-loaded), bigger touch targets (minimum 48px), and a sticky “Post Ad” floating action button. Infinite scroll replaced pagination. We cut the average load time from 4.8s to 1.9s by aggressive image compression and caching.
2. Search & Filters That Actually Work on Phones We moved filters to a clean bottom sheet that slides up with one tap. Added quick preset chips (“Under $500”, “Within 10 miles”, “With photos only”). Real-time search-as-you-type with debouncing so the keyboard does not lag.
3. The Photo Upload Nightmare Fix This was our biggest win. Previously users had to tap “Add photos” → grant permission → select → crop → repeat. We switched to a native-feeling multi-select camera roll picker plus drag-and-drop reordering. Added instant compression + background upload. Average time to post an ad dropped from 87 seconds to 19 seconds.
4. Messaging & Trust Signals We made the chat bubble always visible on listing pages. Added seller verification badges and “Last active” timestamps that are actually accurate. Small things, but they reduced the “is this seller real?” hesitation.
5. Relentless A/B Testing We ran 11 separate experiments over nine weeks using Feature Flags and PostHog. Some ideas flopped hard (the fancy card-flip animation actually increased bounce rate by 14%). Others crushed it.
   We shipped in small batches every Tuesday. No big “relaunch” announcement, just quiet, continuous improvement.

The Results That Made Us Do a Double-Take
After nine weeks the numbers spoke for themselves:
• Mobile bounce rate dropped from 68% to 41%
• Mobile conversion rate jumped from 1.8% to 3.9% (117% increase)
• Average session duration on mobile more than doubled to 2 minutes 14 seconds
• First-time user retention (Day 7) climbed from 19% to 34%
• Weekly transactions from mobile users increased 68%
• Support tickets related to UX fell by 54%
More importantly, our Net Promoter Score on mobile went from 12 to 41. People started saying things like “Finally feels like a proper app” in reviews.
The revenue impact was real. Those extra conversions directly paid for two months of runway we did not expect to have.

What I Learned as a Bootstrapped Founder
The biggest lesson was not technical, it was mindset.
We had been treating mobile as “desktop but smaller.” Wrong. Mobile users have completely different intent, attention span, and context (they are often walking, commuting, or multitasking). Speed beats beauty every single time. One-tap actions beat clever UI every single time.
We also learned that data + real user conversations beats assumptions by a mile. Every change we made was driven by evidence, not opinions.
And most importantly: you donot need a huge redesign budget or a product team of 12. You need ruthless prioritisation and the willingness to ship ugly but functional improvements every single week.

What About You?
If you run any kind of marketplace, SaaS, or consumer app right now, what is your biggest mobile UX headache?
Have you fixed something that felt impossible at first, only to see massive results once you actually shipped it? Drop your story (and your numbers) in the comments. I read every single one and always reply.
Let us learn from each other, because mobile is not the future anymore. It is the present, and the founders who get it right are the ones pulling ahead.