I logged my progress every two business days across my first co-op term, which left me with an unusually complete record to read back: the first weeks learning platforms I’d never touched, the first automation I shipped, the failures in between, and a shift partway through in the kind of work I was targeting. Two things surfaced that I would not have noticed without the record.
What’s worth automating isn’t what’s easiest to automate
My early work targeted monthly reporting, which is the obvious candidate: repetitive, well-defined, and easy to point at. But reading the term back in order, most of the time on my team goes to day-to-day tasks. Those are harder to automate, more error-prone, and carry more upfront cost, which is exactly why they tend to get skipped, and exactly why the payoff is larger when they don’t. The second half of my term shifted toward those, and it changed how I pick problems.
Leading with impact instead of mechanics
My supervisor’s standing question “what is the objective, why are you doing it, and how will it help the team?” changed how I write about my own work. Reading my updates in order, the early ones lead with mechanics and the later ones lead with impact. That’s a measurable change in behaviour, and I only saw it because I had eight weeks of my own writing to compare.
What I’d do differently
I let my objectives emerge over the first few weeks rather than setting them deliberately on day one. It worked out, but the first two weeks had less direction than it needed, and I’d define them explicitly at the start of a future placement.