From 3 decades+ as a software engineer to my first solo Android app (8 days in)
After 32 years working as a software engineer for other people, I shipped my first independent product 8 days ago. It's SO EXCITING!!! And fun, too.
It’s called ClockOn — a simple Android time tracker for freelancers working across multiple clients.
I didn’t start with a startup idea - I built it because I couldn’t find a tool that matched how I actually work:
multiple clients
quick switching during the day
clear weekly totals for billing
no account
no cloud
no ads
works fully offline
So I made one for myself and put it on the Play Store.
I was tempted to call it YATT - Yet Another Time Tracker (after YACC, the compiler-compiler from the 1970s)... probably best I didn't...
Where it is now
It's been live for 8 days now
https://play.google.com/store/apps/details?id=com.roxburgh.clockon
58 downloads and
40 'active' users (still small, but real people using it)
Completely free while I learn how people actually use it
What I’m trying to figure out
My goal is not hypergrowth. I’d like this to become a calm, sustainable product that freelancers rely on and that earns a steady income over time.
Right now I’m watching for:
Do freelancers use it several days per week?
Do they set up multiple clients?
What do they expect from reporting?
The next likely steps are better reporting and data export, but I’m deliberately trying not to overbuild without real usage signals.
I was inspired by Angel Cee:
https://www.indiehackers.com/post/why-indie-founders-fail-the-uncomfortable-truths-beyond-build-in-public-b51fd6509b
"Marketing is not what you do after you build. Marketing is the process of discovering what to build."
The hard part
For most of my career, someone else defined the problem and I delivered the solution.
Now I have to find the users myself — which is a completely different skill.
If you’re a freelancer who bills hourly
I’d genuinely value knowing:
How you currently track time
How it works for you and where it fails
What your weekly review / invoicing workflow looks like
What makes a time tracker “stick” for you long-term
Not trying to sell anything — I’m trying to understand whether I’m solving a real problem in a way that works for other people.
Thanks!
The truth about early metrics
One thing I’ve learned building this app is: question every metric, even when it comes from a reliable source.
In my original post I mentioned ~58 downloads and ~40 active users. It turns out most of those were... me.
When you’re developing an Android app you uninstall/reinstall a lot during testing (especially in the emulator). Each reinstall fires a
first_openevent in Firebase, which is presented as a new user if you're not filtering your data properly.Lesson learned: set up analytics tags and filters early.
Once I cleaned that up, the numbers were a lot smaller — but also a lot more real!
Since then I’ve started sharing the project on social media and something much more valuable has happened: real user feedback.
On Wednesday one user tried the app for real work, gave thoughtful feedback, and asked for a new feature. It was shipped two days later.
That loop — release → engagement → improvement → release — is the most satisfying part of this whole process for me.
The user numbers are still small, but they’re real now. And the conversations are real too.
For now I’m focused on learning from those early users and improving the app. Monetisation can wait a little while longer.
But I can now see something moving... and I know it's not my own tail!
Quick build update (2 weeks after launch):
I’ve just pushed a new version based on sitting next to a few people while they tried to use it for real work and watching where they hesitated.
A few things:
The pre-created account name “Default” made sense to developers but confused almost everyone else. Renaming it to something obviously temporary (“untitled”) immediately reduced friction.
The original UI was slick and clean but too subtle. Features I thought were self-explanatory needed short, well-timed prompts.
Most people start with a single client even if they regularly work for several. The second account may be the real “this fits my workflow” moment — not the third.
So this release is mostly:
– gentle prompts for multi-client use
– removing small points of hesitation
– UI consistency
Along with analytics improvements to track the second account creation rate.
No new headline features.
At this stage, behaviour changes from tiny UX tweaks seem more important than adding functionality.
One thing I’m still trying to understand:
If you’re a freelancer billing by the hour, what makes you trust a new time tracker enough to use it for real client work rather than just testing it?
That feels like the real barrier.