AI ROI is the most-measured, most-exaggerated number in the org. The honest version is uncomfortable — and it's the only one that survives a board meeting.
AI ROI claims are easy to inflate: count the savings, ignore the costs, attribute anything that improved to the AI. The honest measurement is narrower, less impressive, and the only one that holds up when someone asks how you got the number.
The Cashflow Frame: Count Both Sides
ROI is savings minus costs, both sides counted. The savings side: labor replaced, time saved, revenue gained, errors reduced. The costs side: the tool, the integration, the ownership, the opportunity cost, the rework. Most AI ROI counts the first and hides the second.
- Savings: labor, time, revenue, errors.
- Costs: tool, integration, ownership, opportunity, rework.
- ROI = savings − costs, both sides.
The ROI Diligence: Attribution
The hardest part is attribution: did the AI cause the improvement, or did it ride a trend? The honest test is a counterfactual — what would have happened without the AI? Without that, every improvement gets attributed to the AI, and the ROI is fiction.
- Did the AI cause it, or ride a trend?
- What's the counterfactual?
- Is the improvement durable or a launch bump?
What to Cut From the Claim
- Improvements you can't attribute to the AI.
- One-time savings presented as recurring.
- Costs you didn't count (integration, ownership, opportunity).
- The launch bump treated as the steady state.
The Honest Number
The honest number is smaller than the inflated one — and it's the one that survives scrutiny. State the savings, the costs, the attribution, and the counterfactual. If the honest number is positive, ship; if it's not, the AI didn't earn its keep, regardless of how the demo looked.
Conclusion
Measuring AI ROI without lying to yourself means counting both sides, attributing honestly, and cutting what you can't defend. The honest number is smaller — and it's the only one that survives a board meeting or a follow-up question.
About FACTA
FACTA helps startups and growth-stage teams turn AI into production systems that keep running — not demos that impress once.
We design the architecture around the parts that actually break under real usage: tooling you own, credentials you control, failover, cost controls, observability. The boring infrastructure that keeps a system alive after launch.
Led by Matías Baglieri and Carolina Fogliato, we focus on one thing:
AI leadership that builds. Not just advises.
Send us your last AI ROI claim.
We'll tell you what it would take to make it honest. See why AI projects fail at execution for the failure side.
Explore AI Strategy