Sometimes the best tool is the wrong tool
From dashboards to digital art, via one very stubborn idea.
I’ve always enjoyed using tools for things they were never designed to do.
Partly because it’s fun. Partly because I like the challenge. Mostly because I’m curious how far I can push something before it breaks, or does something completely unexpected.
I’ve worked in data for around 15 years, and over that time I’ve used plenty of data visualisation, charts and dashboards tools. The one that always stood out to me was QlikView.
Not because it made pretty dashboards out of the box or because it was the silver bullet in the field (far from it), but because it was ridiculously flexible.
Buttons, sliders, input fields, custom functions… you could build things that felt more like web apps than dashboards. Half the fun was seeing what you could make it do.
Around the same time, I’d become obsessed with flat design.
The problem was that I couldn’t draw.
I also wasn’t particularly motivated to spend hundreds of hours learning Adobe Illustrator just to create the kind of artwork I had in mind.
Then it occurred to me…
What I actually liked about flat design wasn’t drawing.
It was the clean shapes. The bold colours. The simplicity.
And I already knew a tool that was really good at drawing shapes on a screen.
So I stopped trying to become an illustrator and started wondering:
“What if I used QlikView to make geometrical art instead?”
It was such a ridiculous idea that I had to try it.
Building the app ended up being just as enjoyable as using it. Instead of visualising data, it generated abstract compositions that I could tweak and experiment with.
I ended up creating quite a few pieces, and even used one as the cover artwork for a song I released in 2023.
It wasn’t perfect, though.
The gradients weren’t particularly smooth, and the exported images were tiny. To get a usable cover image, I had to export it with exaggerated colours, vectorise it in Illustrator, then manually tweak everything back again. Ironically, the imperfect vectorisation created little artefacts that I actually ended up liking, so they stayed.
Eventually another limitation caught up with me.
QlikView is proprietary software. When my licence expired, so did my art app.
For years I thought that was the end of it.
Then recently I rebuilt the whole thing as a web app with a lot of help from AI.
Now it’s far more responsive, much easier to customise, and exports images at a much higher resolution.
In the end, I never learned to draw.
Instead, I forced a business intelligence tool to make abstract art.
Years later, thanks to AI, it has a second life as a proper web app rather than a forgotten experiment trapped behind an expired software licence.
Have a play, make something colourful, and if you discover a use for it that never crossed my mind… please let me know.
Those are always my favourite stories.
