No clients yet. Just the work, and an honest account of how it was built.
Background
I’m teaching myself data analysis, in Nairobi, working through a structured plan and publishing each project as it’s finished. That’s the whole picture, and I’d rather state it plainly than dress it up.
So I can’t show you a client list. What I can show you is the reasoning behind every project on this site — the method, the checks I ran, and the point where my first approach turned out to be wrong.
That’s deliberate. Experience is the one thing I can’t shortcut, so the useful thing is to make my thinking legible instead — enough that you can judge it yourself rather than take my word for anything.
How I work
Four rules I don’t bend.
None of these are clever. They’re the things that stop a wrong number reaching somebody who’s about to make a decision with it.
Reconcile before you report
A figure that hasn’t been checked against a second source is a guess with a decimal point on it. Every join gets a row count on both sides before anything moves forward.
Say what it leaves out
Every number excludes something — cancelled orders, null dates, the branch that files late. The useful ones tell you what, so nobody builds a decision on a gap they couldn’t see.
Publish the wrong turn
The first attempt is usually the instructive one, and almost nobody shows it. Every write-up here includes the approach I abandoned and the reason I abandoned it.
Sometimes the answer is don’t
Occasionally the honest finding is that nothing here is worth fixing yet, or that the data can’t answer the question being asked of it. That gets said too.
Direction
Analyst now. Engineer next.
Analysis is where I’m concentrating first. Making existing numbers trustworthy is the foundation everything else sits on, and it’s the part I want to be genuinely good at before I move past it.
But cleaning and reporting only hold for as long as nothing underneath them changes, and eventually something always does. So I’m working toward the layer below: transformation modelling, warehouse design, pipelines that hold their shape as an organisation grows. Analytics engineering, in other words — stated as a direction I’m travelling in, not a credential I already hold.
The work says it better than I can.
If you’d rather see how I think than read about it, the projects section walks through a full analysis — including the ranking I built first, which pointed at entirely the wrong sellers.