Skip to content
← Resources

How to measure an AI pilot at work

A simple way to tell whether an AI tool is actually saving time before you roll it out.

Updated 10/15/2026

The short answer Pick one task, measure how long it takes now, try the tool for two to four weeks, and measure again, including the time spent checking the output.

Worked example Before: writing a weekly report takes 90 minutes. With AI: 40 minutes drafting plus 20 minutes checking, so 60 minutes. The real saving is 30 minutes a week, not 50, because checking counts.

What to record Time per task before and after. Errors found during review. Whether the people doing the work would keep using it. Any cost of the tool.

Sources NIST AI Risk Management Framework, "Measure" function. Your own before-and-after timings are the most important evidence.

Limitations Small pilots can be noisy. Run it long enough to cover normal busy and quiet weeks.

Next step The free career check lists your tasks with hours, which gives you a starting baseline.