Automation is only worth it if the hours it saves cost more than the tool. Enter the time you expect to recover, the cost of that time and what the software actually costs, and you get monthly savings, first-year ROI and the payback period.
Use the fully loaded hourly cost (salary plus employer contributions and overhead), not the gross hourly wage. If the saved hours are not really reinvested or reduced, they are not a saving: this calculator measures recovered capacity, not cash in the bank.
How the return is calculated
monthly labour saving = hours × people × hourly cost × 52 ÷ 12annual saving = monthly labour saving × 12first-year ROI = (annual saving − tool cost × 12 − implementation) ÷ (tool cost × 12 + implementation)payback = implementation ÷ (monthly labour saving − monthly tool cost)
ROI is expressed over the first year because that is when the implementation cost hits. From year two onwards the same annual saving is compared only against the subscription, so the return improves.
Worked example
Three people recover 4 hours a week each, at a loaded cost of 30 per hour, with a tool at 150 per month and 1,200 of implementation:
| Concept | Value |
|---|---|
| Monthly labour saving | 4 × 3 × 30 × 52 ÷ 12 = 1,560 |
| Annual labour saving | 1,560 × 12 = 18,720 |
| First-year cost | 150 × 12 + 1,200 = 3,000 |
| First-year net gain | 18,720 − 3,000 = 15,720 |
| First-year ROI | 15,720 ÷ 3,000 = 524% |
| Payback | 1,200 ÷ (1,560 − 150) = 0.85 months |
A payback under three months is usually a comfortable decision. Beyond twelve months, the estimate deserves a second look at how the saved hours will actually be used.
What this model leaves out
- Adoption time: the first weeks are slower, not faster, while the team learns the tool.
- Maintenance, integrations, training and internal support hours.
- Savings that require a hire not made or a role reduced — otherwise the saving is capacity, not cash.
- Error reduction and faster reporting, which are real but hard to monetise.
- Price increases after the first year or minimum contract terms.
- Tax treatment of the subscription and the implementation, which varies by country.
Frequently asked questions
Should I use gross salary or fully loaded cost?
Fully loaded: salary plus employer contributions, benefits, software and overhead. Using gross salary underestimates the value of the time and makes weak projects look profitable.
What payback period is reasonable?
There is no universal threshold, but under six months is easy to defend, six to twelve needs a clear plan for the saved hours, and beyond twelve is a strategic bet rather than a quick win.
Does saved time count if nobody is laid off?
It counts as capacity: the same team can absorb more work, reduce overtime or improve quality. Treat it as a saving only if you can say what that capacity will be used for.
Related tools and guides
How the return moves with the hours saved
Same team (three people at 30 per hour), same tool (150 per month, 1,200 to implement). Only the hours recovered per person per week change:
| Hours saved per person/week | Monthly saving | First-year ROI | Payback |
|---|---|---|---|
| 2 | 780.00 | 212% | 1.90 months |
| 4 | 1,560.00 | 524% | 0.85 months |
| 6 | 2,340.00 | 836% | 0.55 months |
| 8 | 3,120.00 | 1,148% | 0.40 months |
The model is very sensitive to the hours figure, and that is the number people get wrong. Doubling the estimate from 2 to 4 hours does not improve the case by 100%: it takes the payback from almost two months to under one, which is the difference between "worth a trial" and "obvious yes".
Where the saving actually goes
An automation project produces one of four outcomes, and only two of them are money:
- Cost removed. A role, an external service or overtime disappears. This is cash, and it is the strongest case.
- Capacity recovered. The same team does more work in the same hours. This is real, but only if you can say what the extra capacity will be used for.
- Queue reduced. Work that used to pile up gets done sooner. Valuable, but it usually does not show up as a saving in the accounts.
- Nothing. The tool is adopted, the hours are saved, and the time is quietly absorbed by meetings or rework. This happens more often than the calculators assume.
Before signing, decide which of the four you expect and how you will check it ninety days later. If the answer is "capacity", set a concrete destination for it: a project that was postponed, a service you did not offer, overtime you stopped paying.
What a realistic implementation looks like
The payback in this calculator starts counting from the moment the tool works. In practice the first month is slower than before: the team has to learn, the data has to be migrated and some processes have to be redesigned. A more honest version of the calculation adds those first weeks as a cost:
- Setup and migration: the implementation cost, plus the hours of your own team spent on it.
- Learning curve: count the first four to six weeks at partial productivity, not at full speed.
- Integration debt: the connections with other tools are rarely finished on day one.
- Ownership: someone has to maintain the automations, or they break silently and the saving disappears.
How to check it was worth it
Ninety days after going live, compare three numbers with the ones you used to justify the purchase: hours saved per week, tool cost actually paid (with taxes and add-ons), and what happened to the recovered time. If the hours are not visible in any of the four outcomes above, the project did not fail technically - it failed at the destination.
More questions
What hourly cost should I use?
The fully loaded cost: salary plus employer contributions, benefits, software, workspace and overhead. For an employee, a common rule of thumb is 1.3 to 1.4 times gross salary; for a freelancer, the hourly rate you need to charge.
Should I include the tax deduction of the subscription?
It improves the real return slightly, because the cost is usually deductible as an expense. The calculation is deliberately conservative here: it treats the cost as gross, so the real payback is a little better than shown.
Does it work for a one-person business?
Yes, and it is often the clearest case: set people to 1 and use your own hourly cost. Saving five hours a week for a solo founder is usually worth far more than the subscription.
What if the tool never gets adopted?
Then the saving is zero and the cost is real. That is why the pilot matters: run it with one team for thirty days and measure the hours before committing to the annual plan.