How to calculate automation ROI before you build

By the end of this you'll be able to estimate the automation ROI of a workflow before you build it, so you spend your effort on the work that actually pays back.

The mess

Automation gets sold as a feeling. It will save so much time. Everyone nods, something gets built, and six months later nobody can say whether it was worth it. The flashy automation that demos well often saves almost nothing, while the boring one nobody talks about quietly returns an afternoon a week. Without a number, you are decorating, not deciding.

The wrong way people solve it

They automate by vibe. They build the thing that sounds impressive, or the thing a competitor bragged about, or the thing that is technically fun to build. Cost is treated as an afterthought and savings are never measured, so the project is declared a success because it exists, not because it paid for itself. Effort goes to the loudest idea instead of the most valuable one.

How to size automation ROI

Automation ROI is not mystical, it is arithmetic you can do on a napkin. Take the task: how many times does it happen a month, how many minutes does it eat each time, and what is an hour of that person's time worth? Multiply it out and you have the gross annual saving. Subtract what it costs to build and run, including the boring maintenance. What is left, and how fast it pays back, is your automation ROI.

Trigger (a task repeats) → Decision (frequency × time × hourly cost, minus build and upkeep) → Action (build only if it pays back fast) → Human review (did the saving show up?) → Alert (it drifts or breaks) → Record (track the real hours saved).

What I'd build once the automation ROI is clear

I would rank every candidate by payback, not by excitement, and build the one that returns the most hours for the least upkeep first. High frequency, clear rules, low judgment: that is where automation ROI is highest. Then I would actually measure it after launch, not assume it. A workflow that was supposed to save five hours and saves one is telling you something, and you only hear it if you look.

What can break

Counting the hours saved but ignoring the hours spent building and maintaining, so the ROI is fiction. Automating a rare task that never adds up no matter how clever it is. A saving that exists on paper while the freed-up time quietly leaks into other busywork. And measuring nothing, so you repeat the same expensive guess next quarter.

What the business gets

A short, honest list of what to automate first and what to leave alone, ranked by what it actually returns. Effort spent on the workflows that pay back fastest. And the ability to defend an automation decision with a number instead of a feeling, to yourself or to whoever signs the cheque.

If you cannot estimate the hours a workflow saves before you build it, you are not investing in automation. You are gambling on it.

Bring me the task you think is worth automating. I'll help you run the automation ROI before you spend a thing building it.