Ask a management team which process to automate and the answer arrives quickly, usually the one that annoyed someone last week. Ask the people who run it, walk through it step by step and measure it, and a different picture appears. The step everyone complains about is often not the one that costs the most. The process study exists to find out before money is spent.
Map the process as it runs, not as it is documented
The procedure manual describes the intended process. The real one has workarounds, re keying, emails that replace forms and a spreadsheet somebody keeps on the side. Sit with the people doing the work, follow one transaction from start to finish and draw what actually happens: every step, every handoff, every wait.
Measure three things
Cycle time, from the moment the process starts to the moment it finishes, including waiting. Touches, the number of times a person handles the same item. Errors, the share of transactions that come back for correction. These three numbers, measured over a few weeks, show where the time and the cost really go.
Remove waste before automating anything
Automating a step that should not exist makes the waste faster. Re keying data from one system into another is not a step to automate; it is a step to eliminate with an integration. A double approval that never rejects anything is a step to remove. The improved process is usually shorter than the automated version of the old one.
Rank by impact and effort
With the measured process in hand, each candidate improvement can be placed on two axes: how much cycle time, error or cost it removes, and how hard it is to build. The first project is the one high on impact and low on effort. It pays for itself quickly and builds the confidence needed for the harder ones.
Agree the numbers before the build
Write down the current cycle time, touches and error rate, and the target for each. Measure again after go live. This is how an automation project proves its value, and it is how the next one gets approved.
Key takeaways
- Map the process as it really runs, with the people who run it.
- Measure cycle time, touches and errors before deciding anything.
- Remove waste first; automating a step that should not exist makes it fail faster.
- Rank improvements by impact and effort and write down the target numbers.
Want to apply this in your company?
A conversation costs nothing and usually clarifies a lot. We will tell you honestly whether and how we can help.
