How to Calculate ROI on Automation Projects Before You Build
A practical ROI framework for automation projects, including the full formula, real example numbers, and how to account for lead capture improvements and tool costs before you commit to building.
Haroon Mohamed
AI Automation & Lead Generation
Why most automation ROI estimates are wrong
Business owners either overestimate or underestimate the return on automation projects. Overestimation happens when you count every hour theoretically freed up without accounting for how that time actually gets reallocated. Underestimation happens when you only count labor savings and ignore the revenue side — particularly improved lead response time and capture rates.
A rigorous ROI estimate before you build does two things: it tells you whether a project is worth doing at all, and it sets a baseline you can measure against once the automation is live.
This post walks through the complete formula and applies it to a real example.
The formula
Annual ROI =
(Hours saved per month × Hourly cost × 12)
+ (Additional leads captured per month × Close rate × Deal value × 12)
- (One-time build cost + Monthly tool cost × 12)
Let's break down each component.
Component 1: Hours saved
This is the most straightforward piece but it requires honest estimation.
How to estimate it:
For each task the automation replaces, calculate:
- How often the task runs (per day, per week, per month)
- How long the task takes each time it runs
- Who performs it (what is their effective hourly cost, including benefits and overhead if applicable)
Example: A home services company manually sends follow-up texts to every inbound lead. One operations employee spends roughly 45 minutes per day doing this, including opening the CRM, finding the contact, copying the message template, sending the text, and logging the activity.
- 45 minutes/day × 22 working days/month = 16.5 hours/month
- Effective hourly cost of that employee: $22/hour (including overhead)
- Hours saved per month: 16.5
- Monthly value: 16.5 × $22 = $363/month
- Annual value: $4,356
That's real money but it's not transformative on its own. This is where the revenue component changes the calculation.
Component 2: Lead capture improvement
This component is where automation ROI becomes significant for sales-driven businesses. The research on lead response time is consistent: responding to a lead within 5 minutes versus responding within 30 minutes dramatically increases the probability of making contact.
InsideSales.com (now Xant) published research showing that contacting a lead within 5 minutes of their inquiry produces a 100x higher contact rate than contacting them after 30 minutes. The exact multiplier varies by industry, but the directional finding is reliable across contexts.
How to estimate additional leads captured:
"Captured" in this context means leads that entered your pipeline and received timely follow-up — versus leads that submitted a form after business hours and were never contacted before they moved on.
Conservative assumptions for a home services company:
- Inbound leads per month: 200
- Percentage that arrive after business hours (when no one follows up manually): 35% = 70 leads
- Percentage of those 70 that, with an immediate automated follow-up, would engage: 20% = 14 additional engaged leads per month
- Close rate on engaged leads: 15%
- Average deal value: $800
Monthly additional revenue: 14 × 15% × $800 = $1,680/month Annual additional revenue: $20,160
This is a conservative estimate. Real performance depends on your industry, offer, and the quality of your follow-up message. But even at half this rate, the revenue impact dwarfs the labor savings.
Component 3: Build cost and tool costs
Build cost is what you pay to design and configure the automation — whether that's your own time or an external consultant's fee.
For a lead follow-up automation in GoHighLevel:
- Build cost (consultant rate): $500-$1,500 depending on complexity
- Monthly tool cost: GHL is $97-$297/month depending on plan (the automation capability itself is included in the subscription)
For a Make.com-based multi-step automation:
- Build cost: $800-$2,500 depending on complexity
- Monthly tool cost: Make.com plans start at $9/month for 10,000 operations, $16/month for 40,000 operations, $29/month for 150,000 operations
For an n8n self-hosted setup:
- Build cost: higher (requires technical configuration): $1,500-$4,000
- Monthly tool cost: VPS hosting $5-$20/month
Putting the full example together
Scenario: Home services company, 200 leads/month, one operations employee doing manual follow-up
Annual value:
- Hours saved: $4,356
- Additional revenue from improved lead capture: $20,160
- Total annual value: $24,516
Annual cost:
- Build cost (one-time, amortized over 2 years): $1,000 / 2 = $500/year
- Monthly tool cost (GHL already in use, no incremental cost): $0
- Total annual cost: $500
Net annual ROI: $24,516 - $500 = $24,016 ROI percentage: ($24,016 / $500) × 100 = 4,803%
Even with much more conservative assumptions — say, only 5 additional leads captured per month — the ROI is strongly positive in year one.
What this framework misses
Indirect benefits. A well-documented automation creates organizational knowledge that persists even when staff turns over. The value of that knowledge is real but hard to quantify.
Reliability improvements. Humans make errors under repetitive conditions. An automation runs the same sequence correctly every time. Eliminating errors has value, particularly when errors mean missed follow-up with a high-value prospect.
Staff reallocation value. If the 16.5 hours freed per month are genuinely redirected toward higher-value work — more sales calls, better client service, building new offerings — the value created by that reallocation compounds the ROI. Be conservative here: most "freed" hours don't get cleanly redirected unless you actively manage it.
What it misses on the cost side: maintenance. Automations break. APIs change, platform updates break field mappings, webhook payloads change format. Budget 1-2 hours per month of maintenance time for any non-trivial automation, and factor that into your cost estimate.
The minimum threshold worth building
As a rough guideline: if an automation's projected net annual return is less than $2,000 or takes more than 3 months to recoup the build cost, the project is a low priority unless it has strategic value beyond the direct financial return (e.g., reliability, scalability, documentation).
Most lead follow-up, intake, and onboarding automations for service businesses clear this threshold easily. The projects that don't are typically low-volume, one-off tasks that don't repeat often enough to justify the build time.
Sources
- Xant (formerly InsideSales.com): Lead response time research on contact rate improvement
- GoHighLevel pricing: gohighlevel.com/pricing
- Make.com pricing: make.com/en/pricing
- n8n.io: self-hosted setup documentation and cloud pricing
Want help running this calculation for your specific workflows before committing to a build? Let's talk.
Need This Built?
Ready to implement this for your business?
Everything in this article reflects real systems I've built and operated. Let's talk about yours.
Haroon Mohamed
Full-stack automation, AI, and lead generation specialist. 2+ years running 13+ concurrent client campaigns using GoHighLevel, multiple AI voice providers, Zapier, APIs, and custom data pipelines. Founder of HMX Zone.
Related articles
Automation Stack Architecture: The 5 Decisions That Determine Whether Your Stack Scales
When operators talk about their automation stack, they usually focus on tools: "we use GoHighLevel and Make.com." Tools are real, but they're a tactical layer. The strategic layer is architecture — t…
The Founder Time Allocation Trap: Why You Keep Doing $20 Work When You Should Be Doing $200 Work
If you've been running a service business for more than a year, you've had this thought: "Why am I doing this? Someone else should be doing this." The thought hits while you're updating a CRM record …