How to choose the first process so automation pays off from the start
When a company decides to automate, the temptation is to start with the biggest process, or the one that comes up most in meetings. It often makes sense to do the opposite: start with a contained one that repeats a lot and has clear rules.
If you've already done the sum of the time that goes into manual work, you have the list. Now it's time to choose.
What a good first candidate looks like
- It repeats often. Every day or every week. Something done once a year rarely justifies the effort.
- It follows clear rules. If two people would do it the same way, a machine can too. If it depends on judgement in each case, it's better left for later.
- The data is already digital. Emails, PDFs, spreadsheets, a system. If everything is on paper, that needs solving first.
- The result is visible. Hours that come back, errors that disappear, payments that arrive sooner. A visible first result helps make the case for the next one.
- If it fails, nothing serious breaks. For the first project, pick something where a mistake is easy to fix.
Examples that usually tick every box
- Entering incoming invoices and receipts: reading them and moving the data into the system. That's what we did in a case we shared in February, where operating time fell by 60%.
- Sending payment reminders when an invoice falls due.
- Building the weekly report from data already in the system.
What to leave for later
- Processes that change every week.
- Decisions that need human judgement.
- Processes nobody fully understands. Automating something disorganised just makes it disorganised faster; it needs sorting out first.
Measure before and after
Before you start, write down how long it takes and how many errors there are. After a month, measure again. That comparison is what justifies the next step.
Have you spotted the first process you'd like to automate? Get in touch and we'll review it together.




