Year in Review 2018
Hi! In case you haven't been here before: fman is a cross-platform file manager. You use it as an alternative to Explorer on Windows, or Finder on Mac. It lets you browse directories, copy and move files, etc. All this much more efficiently than the "normal" file managers provided by the operating system. In case you know Sublime Text and Total Commander: fman essentially combines the best ideas of the two. Here is what it looks like:
One year ago, fman had had some respectable successes. It was selling its first licenses. People were already loving it. But fman was also still missing some very basic features:
- fman didn't have a context menu.
- It was not possible to cancel Copy/Move operations.
- Performance was pretty terrible.
All these things were fixed or at least greatly improved in the past twelve months. After an interesting discussion, fman received a scriptable but non-native context menu. It now also shows cancellable progress dialogs for file operations. Finally, performance was hugely improved as well: A year ago, it took fman an unacceptable 18 seconds to display a folder with 5,000 files. This is now down to under a second. This 20x speedup required a lot of work, and it's still not as fast as it should be. But it's a start.
fman's business metrics have improved as well: At the end of 2017, fman was selling around $350 per month in licenses. Nowadays, roughly one license is sold per day, for a monthly revenue of ~ $500. Retention – that is, the percentage of people who use fman a second time within the first week of initially downloading it – increased from 20% to 30%. If we were talking about bigger numbers, this increase would translate almost directly to more license sales, and is thus highly significant.
Nevertheless, both of the above improvements fall short of the goals I had set for myself in the post about 2017: I had wanted to double revenue to $700 and increase retention to 40%. While it's humbling that every day there is someone on this world who buys fman, the number $500 is not a sign of a great business.
Another goal for 2018 was to open source fman's build system. This was born
out of the frustration that months of fman's development time went into
tasks completely irrelevant to its mission: Taking fman's Python source code
and turning it into a standalone executable, say an .exe on Windows, is
already non-trivial. Then creating an installer on each operating system:
fmanSetup.exe on Windows,
fman.dmg on Mac, etc.
And finally, implementing automatic updates. These tasks, however mundane
they are, require learning and integrating a lot of different technologies.
Now that fman's build system is
on GitHub, other developers
will hopefully not have to waste the same months as I.
So what's next? Here's what's most important right now in my eyes:
- Custom Theme Support because it's the most-requested feature.
- An API that lets fman plugins implement custom columns. This achieves several much-requested features: 1, 2, 3.
- Further improvements to fman's performance and stability.
Have a great 2019!