Start with the process people actually use.
The written procedure and the day-to-day workflow are rarely identical. I start by tracing where the information comes from, who touches it, which rules repeat, and where work tends to stall.
That makes it possible to automate the routine steps without burying exceptions or quietly making a decision the system was never qualified to make.
Good candidates for automation
- Files that are renamed, sorted, or routed using the same rules.
- Weekly and monthly spreadsheets that always need the same cleanup.
- Follow-up lists assembled by copying information between systems.
- Intake and onboarding work with predictable review steps.
- Reports built from the same collection of exports and source files.
What I do not automate blindly
If a step depends on incomplete information, an unusual document, or someone making a judgment call, I keep a review point in the workflow. A visible exception is safer than a clean-looking wrong answer.
What the finished workflow should provide
- Fewer copy-and-paste steps.
- A clear owner for anything that needs follow-up.
- Consistent files, records, and reports.
- An obvious place to review errors and exceptions.
- A tool that fits the existing operation instead of fighting it.
Still rebuilding the same spreadsheet or follow-up list?
Send me the rough version of the process. I can help identify what should be automated and what should stay in human hands.