Analytics and automation
What can we stop doing by hand?
A lot of skilled time goes on gathering and reformatting information. I like finding the parts that can be simplified, then automated.
A note on this piece
This is part of my day-to-day work rather than a single project, so I’ve written about how I approach it. Specific examples are best talked through in person, so they aren’t published here.
01Where it starts
The situation
Much of the work in planning and operations is assembly: gathering, reconciling and reformatting information before anyone can start to think about it.
02What makes it hard
Why it’s tricky
Data lives in lots of systems with different owners and different definitions. Manual processes hold unwritten rules in people’s heads. And automating a confusing process without simplifying it first just makes it confusing faster.
Automate the process you’d design, not the one you inherited.
03My remit
My part
The analytics and compliance teams sit within the organisation I lead. My part is deciding where effort goes and making sure improvements actually get used, not just built.
04How it gets handled
How I go about it
I start with the decision or the recurring task, not the tool. Simplify first, then automate, then check whether time or errors really fell. And I involve the people who run the process from the start, because adoption is where most improvement quietly fails.
05The lesson
What I’ve taken from it
Curiosity about technology only helps when it’s tied to a real problem. It’s also a big reason I build things in my own time: it keeps my sense of what’s easy, what’s hard and what’s worth doing grounded in practice.
Where this sits in the sketch
- Performance
- Execution
Next
DRGNFLY Studios