Have you ever worked on something with the distinct feeling that nobody is going to notice the feature you're pouring your sweat and tears into?
It just happened to me. This afternoon.
I'm working on some data visualisation components. They look good, the API calls are fast, and the experience on the dashboard is already great. We added some nice filters, and we're confident the data we're surfacing is genuinely useful. Most of the time, this would be the end of it. The feature works, the numbers match, the acceptance criteria are met. Ship it.
This time, something slightly different happened.
With tasks like these, I like to keep a final catch-all ticket for small adjustments: polishes we spot along the way, UI bugs that slipped through until the end, and feedback from any side of the team. I was working through some of those items when I noticed something that bothered me as a user.
In the interface, you can set the data granularity to match what you need. It auto-detects the date range and adapts, but you can always tweak it manually. I was checking that this mechanism ran smoothly when I realised that the charts with dates on the x-axis always showed a full date (obviously) as the label.
That works fine with days or weeks. But set the granularity to months, quarters or years, and the label becomes less and less meaningful.
The fix wasn't hard. I added a small utility that reads the global granularity and formats the label accordingly. Just the year when you're comparing years, a concise notation for quarters (Q3 2026), or month and year otherwise. It didn't take long, but it was an extra nobody asked for, squeezed into an already hectic day close to the delivery date. Was it worth it?
I honestly don't think anyone will ever contact support to say they love how the dates change dynamically. I don't think a single user will consciously notice them at all. And that's exactly why I'm sure it was the right call.
"Good UX goes unnoticed, bad UX gets noticed immediately" is a tired cliché, but it's true. In this case nobody will notice, and everyone will just read the charts as expected. But I'm certain that showing a full date would have confused some users at some point, and at the very least made the whole experience a little worse.
American sports have this concept of intangibles. The things that matter enormously without ever showing up in a stat line. I think that in an era where producing code is cheap, intangibles matter more than ever. They're the small polishes that make the overall experience better. The user won't notice, most of my colleagues probably won't either, but that isn't the point. I don't code for recognition. I code for the love of the craft and the product I'm putting out.
Now, before you start garnishing everything in sight: I only shipped this because the scope was genuinely tiny, the blast radius minimal, and the cost to my resources manageable. I wouldn't have gone ahead on a whim if it touched important architecture, or if the effort massively outweighed the benefit. That's just knowing how to work in a team.
But with that caveat in place, I'd encourage everyone to stay alert to these small chances to make the online world a little better. It's the quiet touches that make a product feel more premium, more considered, built with more care. An extra couple of hours here and there feels like a small price to pay.