An Automation ROI Calculator for Small Businesses (Without the Spreadsheet Theatre)

Automation ROI for a small business is simple math the moment you stop trying to do it by gut. Three honest inputs, two cautions, and the payback math falls out. Most workflows that meet the weekly-repetitive-predictable filter clear the bar inside a single quarter, and the ones that do not are usually obvious for other reasons.

The three honest inputs

First, hours saved per week. Measure this before the build, not after, by literally timing the manual process for two weeks. Estimates from memory are always wrong; they are usually wrong on the low side. Second, the loaded hourly cost of whoever does the work today, salary plus benefits plus opportunity cost, divided by working hours per year. For a founder, the opportunity cost number is the right one, not the salary equivalent. Third, build cost plus annual operating cost: agency fee plus SaaS plus infrastructure plus your own time during the build.

Plug them in. (Hours saved per week × loaded hourly cost × 52) − (build cost + annual operating cost). If the result is positive in year one, build it.

A worked example

A founder spending five hours a week on lead triage at a €200 loaded hourly rate is burning €52,000 per year on that one task. A €15,000 build plus €3,000 a year of running cost pays back in roughly four months. After that, every year is pure return, €34,000 in year one, €49,000 in years two onwards.

The same math at a lower hourly rate or fewer hours saved still works as long as the build cost is honest. A €5,000 build that saves three hours a week at €60 an hour is €4,360 in year-one return, positive but tight; the project is worth doing only if the soft factors are real.

The two cautions

First, hours saved is real value only if the time gets reallocated to higher-leverage work. A founder who saves five hours and spends them refreshing a dashboard has not captured the ROI. Build into the engagement an explicit answer to 'what does the operator do with the saved time?' If the answer is vague, the ROI is theoretical.

Second, soft factors, fewer errors, faster customer response, less burnout, are real value that does not fit neatly in the formula. Do not double-count them, but do not ignore them either. The honest pattern: run the hard math first, decide on the basis of that number, treat the soft factors as the tiebreaker rather than the case.

When the calculator says no

Some workflows look like they should automate but do not pay back. Bespoke client work is the most common, every instance is different, the build does not amortise. Quarterly or annual processes, the ROI is real but small, and the build cost dominates. Anything that requires judgement on novel inputs, the LLM gets it wrong with confidence, and the cost of correcting bad outputs eats the savings.

The honest answer in those cases is not to automate. The right move is to streamline the manual process, train the operator, and revisit when the volume changes.

Where to read more

The answer page on automation ROI covers the same math in tighter form. The answer on whether automation is worth it for small business covers the threshold question.

If you want a second opinion on a specific automation candidate, send a short note describing the process and the hours involved. We respond within one working day.