Falnor / Tools
Automation ROI calculator
Put in the work you want to automate and what it would cost. See the hours it frees, how long it takes to pay back, and how the answer changes if you are wrong about how much can be automated. Free, no sign-up.
Book a discovery callThe work you want to automate
Wages plus employer costs.
Be cautious. The rest stays with people.
Rework, refunds, delays. Use 0 to leave errors out.
What it would cost
Hosting, usage fees, upkeep, someone checking it.
The starting numbers are examples to show how the tool works. Replace every one with your own.
What it returns
Check this £23,520 first-year return with usPays back in
4.7 months
Net £3,210 a month once running.
Staff time freed a month
105 hours
About 0.7 of a full-time person (160 hours a month).
Gross saving a month
£3,710
£3,150 time, £560 fewer mistakes.
Net over 12 months
£23,520
After the build cost and running costs.
Net over 36 months
£100,560
670% return on the build cost.
Worth checking against the real work
A 30-minute discovery call goes through these inputs with you, checks them against how the work really runs, and tells you plainly if it is not worth building. Your numbers are filled in for you.
Cumulative position over three years
- Net position after build cost
The line starts below zero at the build cost and rises by the net monthly saving. It crosses zero at the payback point.
How sure are these numbers?
Cautious: 50% automated
7.0 months
Net £10,800 over 12 months.
Your figure: 70% automated
4.7 months
Net £23,520 over 12 months.
Upbeat: 80% automated
4.0 months
Net £29,880 over 12 months.
If it still pays back in the cautious case, the case is sound. This is an estimate from the numbers you enter, not a quote. Real projects differ once the work is mapped.
What it measures
Two kinds of saving, both from the numbers you give: the staff time a system would take over, and the mistakes it would stop. It then takes off the monthly running cost and compares the result with the one-off build cost.
- Hours saved a month = runs a month x minutes each x share automated, divided by 60.
- Time saving = hours saved x cost of an hour. Mistake saving = mistakes a month x cost of one x share automated.
- Net a month = time saving + mistake saving - running cost. Payback = build cost divided by net a month.
- The 12 and 36 month figures are net a month times the months, less the build cost.
Why there are three cases
The share you think a system can do is the least certain number, and it moves the result more than any other. So the page also shows a cautious case (20 points lower) and an upbeat one (10 points higher). If the cautious case still pays back, the case is sound. If it does not, find out why before you spend.
When a build does not pay back
Small or rare tasks. If the total time a task takes across your team is only a few hours a week, a custom build is unlikely to pay back and tidier use of the software you already have is the cheaper fix. Work that changes every time, or needs judgement on each case, also automates poorly. Both show up here as a long payback or none.
Running costs matter as much as build cost. A system that needs a person to check it, a licence, hosting and usage fees all belong in the monthly figure, and should be set honestly.
What to do with the result
Use it to decide whether a process is worth a conversation, and to give a colleague the numbers behind the case. Use the what to automate first tool to rank several processes, and the invoice processing cost calculator if the work is accounts payable. For how we build and keep a process under your control, see business process automation.
Questions people ask
What is a good payback period for automation?
There is no single answer. Many businesses look for payback inside one to two years, but it depends on the cost of capital, the risk and what else the money could do. Check the cautious case as well as the expected one.
Why are the staff cost and error cost separate?
Time saved and mistakes avoided are different benefits and have different evidence. Keeping them apart lets you leave the mistake cost at zero if you cannot support it.
Does it count the cost of changing how people work?
No. Training, a period of running both ways and the time to review results are real. Put them in the build cost or the running cost.
Are the starting numbers real averages?
No. They are examples so the page shows a result straight away. Replace every one with your own figures.
Is this a quote?
No. It is an estimate from what you enter. A real project is priced once the work has been mapped.