How to Calculate the ROI of Business Automation Before You Build Anything
Most companies decide whether to automate based on gut feel. The ones who get the best outcomes build a simple financial model first. Here's exactly how to do it.
Before you spend a dollar on automation, you should be able to answer one question: what is this worth? Not in vague terms — 'it'll save us time' — but in hard numbers. What does the current process cost per week? What will the automated version cost to build and maintain? How many weeks until it pays for itself?
Most businesses don't do this math before committing. They buy a tool, hire a consultant, or kick off a project on the assumption that automation is good and therefore worth it. Sometimes they're right. But without a financial model, there's no way to prioritise which automation to build first, whether the project scope is reasonable, or when it's actually delivered value.
Step 1 — Calculate the current cost
Start with labor. Identify everyone who touches the process and how much time they spend on it per week. Multiply their hourly fully-loaded cost (salary plus benefits, typically 1.25–1.4× base salary) by the hours. This is your direct weekly cost.
- Direct labor: hours × fully-loaded hourly rate
- Error correction: time spent fixing mistakes the process produces
- Downstream delays: cost of decisions or actions that wait on this process
- Opportunity cost: what the people doing this work should be doing instead
Add these up. In our experience, the true cost of a manual process is typically 2.5–4× the direct labor cost once you include downstream effects.
Step 2 — Estimate the automation investment
The cost of automation has two components: build and maintain. Build is a one-time cost — the time and money required to design, develop, and test the system. Maintain is an ongoing cost — the time and money required to keep it running, update it as the business changes, and monitor it for failures.
“A build cost without a maintenance estimate is an incomplete number. Every automation you build is a system you're committing to maintain indefinitely.”
Step 3 — Build the payback model
Divide the build cost by the weekly savings (current cost minus maintenance cost). This gives you the payback period in weeks. A payback period under 26 weeks (6 months) is excellent for most business automations. Under 52 weeks (one year) is acceptable. Beyond that, the automation needs to be delivering significant non-financial value — like error elimination or compliance — to justify the investment.
This model doesn't need to be precise. It needs to be directionally correct. An automation that costs £15,000 to build and saves £3,000 per month has a 5-month payback period — that's worth building. An automation that costs £40,000 and saves £1,000 per month has a 40-month payback — that's worth questioning.
Why this model changes prioritization
When you build this model for five or six automation candidates, something useful happens: the ranking becomes obvious. The process that feels the most painful isn't always the one with the best ROI. The process that runs 50 times a day beats the one that runs once a week, even if the per-run cost is lower. The model lets data drive the sequencing rather than whoever is loudest in the room.
In our Discovery Audit, we build this model for your top three opportunities — including the costs most clients hadn't considered before. You leave with a prioritized list and hard numbers to take to whoever approves the investment.
See what this looks like in your operations.
The Discovery Audit is free. 45 minutes, a written report of your top 3 opportunities, delivered within 48 hours.