Measure before you automate
The question you cannot answer later
Six months after any automation ships, somebody asks whether it was worth it. If nobody measured the task before the machine took it over, that question has no honest answer. You are left with it feels faster, which convinces nobody, least of all the person who signed the check.
The baseline has to be captured before the change, because afterward the old cost is gone and memory always shrinks it.
A week of tallying is enough
This does not require time-tracking software or a consultant. For one normal week, the person who does the task keeps a tally: each time they do it, roughly how long it took, and one note when something made it slower than usual. Five entries a day on a sticky note is a perfectly good dataset.
Multiply out the week and you have the number that matters: hours per month this task currently costs. Write it down somewhere it will survive.
What the baseline buys you
That one number does three jobs:
- It ranks the queue: a task eating twenty hours a month outranks one eating three, whatever anyone's instincts say
- It sets the bar: an automation that returns most of those hours is working, one that returns a sliver is not worth its upkeep
- It ends the worth-it argument before it starts, because both sides are looking at the same number
Measure the after, too
The honest version of the exercise repeats the tally a month after the automation ships, counting the new costs: reviewing the machine's output, handling its exceptions. The gap between the two numbers is the real return, in hours, which is the only ROI metric we trust. Everything else is a demo.
The audit answers this for your business
Two weeks, $2,500 flat ($1,000 for the first three clients), and you get the map of your own automatable work with dollars on it.