Lesson 1 of 5, 11 min with the exercise
Get your before number
Have something to compare with. You can't recover this later.
Measure the current process before you change it. Once the new system is live, you'll be relying on people's memory of how long things used to take, and memory flatters.
Time the task on 10 to 20 real cases, with the people who normally do it. Count how often work has to be redone too, because a faster process with more errors may be no improvement at all.
See it in practice
Invented example: the team says a task takes "about 5 minutes". Timing 15 cases gives an average of 11, because they forgot the lookups and the interruptions. Had they trusted the estimate, the new system would have looked far less useful than it was.
Common mistake: Timing only the easy cases, or only your fastest person. Measure a normal mix with the people who usually do the work.
Try it
Pick your task and time it this week. A shared spreadsheet with a start time, end time and notes column is enough.
You're done when
- You've timed 10 to 20 real cases
- You've recorded the rework rate
- Results are saved where you can find them later