• DESIGN FOR FOUNDERS · 03
If users can't find a feature, you are paying for something that doesn't exist
Abhinav Dhankar · 3 min read
Earlier this year I did a small redesign exercise on WhatsApp's UPI payments. It was self-initiated, nobody at WhatsApp asked for it. The fix I landed on was so small that I almost didn't post it. But the problem behind it is one I now see in almost every product I review.
The feature was there. The starting point wasn't.
WhatsApp is one of the most used apps in India, and it supports UPI payments. But it is a messaging app first, and that shows in how its payment features are placed. When I looked at the home screen, the option to scan a QR code sat in a place that felt accidental, not intentional.
Now think of the person standing at a shop counter who opens WhatsApp only to pay. The feature exists, it works, someone built and tested it. But she has to hunt for where it starts. And the moment a user has to hunt, you have already lost her to the app she knows.
My fix was simple. Move the scan QR icon to the bottom navigation bar and give it a label. Visible. Labelled. Found in one look.
Why this matters more to a founder than to WhatsApp
WhatsApp can afford a buried feature. People will open it a hundred times a day anyway. Your product doesn't have that luxury.
Every feature you ship has a cost: the developer's weeks, the testing, the support articles, the announcement. If users don't find it, all of that cost is still on your books, but the value never shows up. Then the usage report says nobody wants the feature, and the team concludes it was a bad idea. Most of the time it wasn't a bad idea. It was a good feature without a front door.
Built and found are two separate jobs. Engineering does the first one. Design does the second.
• TRY THIS ON YOUR OWN PRODUCT THIS WEEK
Pick the last feature you shipped that didn't get the usage you expected. Open your product as a new user would, and without searching, try to reach it from the home screen. Count the clicks, and note whether anything on the way told you it existed.
If you had to know where it was to find it, you don't have a feature problem. You have a discoverability problem, and that is usually a much cheaper fix than building something new.
For a feature to be useful, it first has to be found.