How to calculate the ROI of AI automation for your business
Before investing in automation, you need a number. Here's the framework I use with every client to calculate ROI — what to measure, how to estimate it, and what payback period to expect.
- roi
- automation
- small business
- guide
- cost savings
The question every business owner should ask before committing to an automation build is: what’s this actually worth?
Not “this sounds useful” — but a specific number. Hours saved per week, revenue recovered per month, cost reduction per year. Compared to the build cost and the ongoing maintenance cost, what’s the return?
Here’s the framework I use with every client.
The three sources of automation ROI
Automation ROI comes from three places: time savings (tasks that were done manually now run automatically), revenue recovery (leads, customers, or money that was being lost that is now captured), and cost avoidance (a function that would otherwise require a hire or an additional service).
Most automations produce returns in more than one category. An appointment reminder system saves the time someone was spending manually confirming appointments AND recovers revenue from no-shows. A lead follow-up system saves the time someone was spending on manual follow-up AND recovers revenue from leads that were going cold.
Calculate each separately, then add them.
Calculating time savings
Step 1 — Measure the actual time. Before relying on estimates, track the workflow for one week with a timer. Administrative time is consistently underestimated because the tasks are distributed across the day in small increments. A business that estimates “maybe 30 minutes per day on intake emails” typically measures 75–90 minutes when they actually track it.
Step 2 — Apply the fully-loaded cost of that time. The right number here is not the hourly wage of the person doing the work — it’s the fully-loaded cost, which includes benefits, overhead, and the opportunity cost of what that person could be doing instead. A reasonable rule of thumb: 1.3–1.5× the base hourly rate for employees, or your own hourly rate for work you’re doing personally.
For a business owner doing the administrative work themselves at an opportunity cost of $75–$150/hour, even 3 hours per week of saved time represents $11,000–$23,000 per year.
Step 3 — Apply the realistic reduction percentage. Automation rarely reduces a task to zero — there are always exceptions, edge cases, and situations that require human intervention. A realistic estimate for a well-built automation is 70–85% reduction in human time for that workflow, not 100%.
Time savings annual value = (Hours per week saved × 52 × Hourly cost) × 0.75
Calculating revenue recovery
Revenue recovery is often the larger number, particularly for automations that address lead response, no-shows, or lapsed customer reactivation.
Lead response automation: How many new inquiries do you receive per week? What percentage currently don’t hear back within 24 hours (estimate honestly — this is usually higher than business owners expect)? What percentage of those are lost to competitors? What’s the average revenue per conversion? That math gives you the annual lead loss you’re experiencing. The automation recovers some percentage of that — typically 20–40% of what was being lost.
No-show reduction: Number of appointments per week × no-show rate × revenue per appointment × annual weeks. Multiply by 0.35 (typical no-show reduction from a properly built reminder sequence) to get the annual recovery.
Invoice aging improvement: Annual billings × (current average days-to-payment – projected average days-to-payment) / 365 × annual cost of outstanding receivables. For a business billing $600K annually, moving from 55-day to 30-day average collection improves working capital by ~$41,000 — which has a real cost-of-capital value even if it’s not direct cash recovery.
The cost side of the equation
Build cost: The one-time fee to design, build, and test the automation. Ranges from $3,000–$80,000 depending on complexity.
Annual maintenance cost: The monthly retainer or hourly maintenance cost, annualized. Typically $400–$1,200/month for systems that include active monitoring and updates.
Annual tooling cost: The cost of the automation platform and any APIs used. Typically $500–$2,000/year for a moderate-complexity system.
Total cost = Build cost + (Annual maintenance + Annual tooling) × Project lifetime in years
Putting it together
A worked example: a home services company in Marietta with 60 appointments per week and a 12% no-show rate.
- No-show revenue recovery: 60 appointments × 12% × $185 average service × 0.35 (reduction) × 52 weeks = $24,055/year
- Time savings (manual reminder sending, 45 min/day): 0.75 hrs × $45 fully-loaded × 260 workdays × 0.80 reduction = $7,020/year
- Total annual value: $31,075
- Build cost: $7,500
- Annual maintenance + tooling: $2,400/year
- Payback period: ~3.5 months
- 3-year ROI: 270%
That’s a typical profile for this type of automation. The payback is fast and the multi-year return is substantial.
When the numbers don’t work
Run this calculation honestly before committing to a build. If the time savings are small (fewer than 2 hours per week) and there’s no revenue recovery component, the ROI on a $5,000–$8,000 build may be poor — 3–4 year payback periods mean the system may need to be replaced before it pays off.
In those cases, there may be simpler solutions: a better-designed intake form, a Calendly link with built-in reminders, a template email that takes 2 minutes to send. Not every inefficiency requires custom automation.
If you want to run this calculation for your specific business before making a decision, book the free 30-minute audit. I’ll map your current workflows, estimate the time and revenue impact, and give you a clear ROI projection you can actually use to make the decision.