How to Build an Automation Business Case Your CFO Will Approve

TL;DR: A CFO approves an automation business case when it is short, conservative, and framed in cash. Name the process, quantify the fully-loaded hours saved, subtract realistic build and maintenance costs, and state a payback period. Lead with net annual cash impact and payback, not a big ROI percentage, and count only hours you can genuinely reallocate or avoid hiring for.

Most automation proposals are rejected not because the idea is bad, but because the case is built for enthusiasts rather than for finance. It leads with a dramatic percentage, hides the cost side, and rests on numbers nobody can trace. This guide gives you a structure your CFO will recognize and trust.

What does your CFO actually care about?

A CFO reads your proposal looking for three things, in this order: is the saving real cash, how long until we get our money back, and can I trust the assumptions. Everything else is secondary.

That changes how you write the case. A finance leader is trained to discount optimism, so a headline like "300% ROI" triggers scrutiny, not approval. A line like "recovers €31,000 in avoidable labor cost per year, payback in 7 months, based on measured task volumes" earns a second read. Speak in the language of cash flow and risk, and you are already ahead of most proposals that cross their desk.

What structure should the business case follow?

Keep the decision on one page. Put the detail in an appendix. The one-pager should contain seven elements:

1. The problem, in one sentence. Which process, how often, and why it is expensive now.
2. The proposed change. What gets automated and what stays human.
3. Hard savings. Cashable hours per year, valued at fully-loaded cost.
4. Total cost. Build, run, maintenance, exceptions, and monitoring.
5. Net cash impact and payback. Year-one net and steady-state net, plus months to break even.
6. Scenarios. Conservative, expected, and optimistic.
7. Risks and assumptions. Stated plainly, with the biggest one flagged.

If any element is missing, expect a question, and unanswered questions are how proposals die in committee.

How do I calculate the savings credibly?

Two rules keep finance on your side. First, value time at fully-loaded labor cost, not salary alone, because an employee costs benefits, tools, workspace, and overhead on top of pay. A common multiplier is 1.25 to 1.4 times base salary. Second, count only cashable time: hours you can reallocate to revenue work, avoid hiring for, or stop paying as overtime. Time that simply evaporates into a less busy afternoon is a benefit, not a saving, and your CFO knows the difference.

For the full formula and a step-by-step calculation, see our guide on the real ROI of automating repetitive work. The short version: annual value of time saved, minus annual automation cost, divided by that cost, with payback tracked separately.

What does a worked one-pager look like?

Here is a compact example for an invoice-entry process, the kind of figures you would present in the appendix and summarize on the front page.

Notice the case still works in the conservative column. That is the point. When your worst-case scenario is positive, the CFO is being asked to approve an investment that pays off even if things go moderately wrong. That is a very different conversation from approving one that only works if everything goes right.

Which costs do people forget, and finance never does?

The savings side gets the excitement; the cost side gets the surprises. Build these in before someone in the room raises them:

• Maintenance. Systems and interfaces change, and someone must keep the automation working. Budget 20 to 35 percent of build cost per year. This is the single most common gap in weak cases, and it is especially large for brittle, screen-scraping automations, as we explain in RPA vs. AI automation.
• Exception handling. No automation reaches 100 percent. Fund the human time that handles what it cannot.
• Change management. Training, documentation, and a temporary dip in output while the team adjusts. Getting this wrong stalls the savings, which is why change management for automation deserves its own line.
• Monitoring. Someone must notice when an automation fails quietly, or you get errors at scale.

A useful discipline: if the case only works assuming full automation and zero upkeep, it does not really work.

How do I handle benefits I cannot put in euros?

Faster cycle times, fewer errors, better employee experience, and capacity to grow without hiring are all real, but they resist a clean number. Do not inflate the headline with them and do not invent figures. List them separately as qualitative support so the CFO sees the full picture while the hard calculation keeps its credibility. Finance will trust a modest, defensible number far more than an impressive one it cannot verify.

How does "pay only when you save" change the pitch?

Traditional automation front-loads risk: you pay to build, then hope the savings appear. A pay-only-when-you-save arrangement inverts that. Because the provider is paid from realized savings, the build-cost line drops toward zero and the payback period effectively collapses. For a CFO, that reframes the decision from "approve a capital outlay and trust the forecast" to "approve a change that only costs us once it demonstrably works." It removes the risk that makes finance cautious in the first place, but it does not remove the need for honest measurement of the hours saved.

How Espai.AI helps

The weakest part of most business cases is the phrase "we estimate." Espai.AI removes it. It silently records desktop and system events, and its AI identifies exactly which repetitive tasks consume time and how much, so your savings line rests on observed data rather than guesswork. That data stays on your own systems and is never seen by humans. Because pricing is pay-only-when-you-save with no upfront cost, the build-risk that sinks most proposals largely disappears, which makes the CFO conversation dramatically easier. See how the model works on the pricing page, or explore the measurement approach in the live dashboard demo.

Key takeaways

• Lead with net annual cash impact and payback period, not a headline ROI percentage.
• Keep the decision on one page, with the detailed calculation in an appendix.
• Value time at fully-loaded cost and count only cashable, reallocatable hours.
• Show the full cost side, including maintenance, exceptions, and monitoring, and present a conservative scenario that still works.
• Tie every number to observed data or a clearly labeled assumption, because unsupported estimates are the fastest route to rejection.