The Evidence Map for Enterprise AI ROI Measurement
The action layer behind the core verdict: how to turn the briefing into a sponsor-ready decision without overstating the evidence.
Use this as a value-realization map before an AI initiative goes to a sponsor, steering committee, or board. The point is not to kill experiments. The point is to stop soft measures from being promoted into hard-dollar claims before finance has a way to recognize them.
First moves before hiring anyone
Ask the controller to define the before state, counterfactual, denominator, and ledger line. If the AI team owns the baseline, mark the result as an internal signal.
Efficiency plays need hard-dollar discipline. Positioning bets need option-value framing and explicit exemption from quarterly ROI promises.
When hours saved are claimed, separate cost actually removed from time reinvested into more work. Most productivity claims die in that second column.
Governance, system integration, change management, token spend, security review, and support all belong in the business case before approval.
Compare the client against its own full cohort of attempts, not industry averages that exclude failed or abandoned deployments.
Owner, briefing, proof
Owner
Finance-owned baseline and value-recognition rule, not an AI-team estimate.
Briefing
Efficiency or positioning decision brief, with the measurement standard named before the pilot starts.
Proof
A ledger bridge that separates task gain, workflow change, cashed savings, redeployed capacity, and shadow costs.
Start with one AI initiative, one finance baseline, and one proof path. If the gap is material, widen to a readiness look at AI value measurement, and build the operating cadence only when the sponsor wants it run.
Claim ledger
- A peer-reviewed study causally identifies firm-level AI ROI.
- The DellAcqua/METR heterogeneity pattern is contradicted or fails to replicate.
- A primary source corrects, replicates, or retracts the MIT NANDA 95% figure.
- An independent audited enterprise-ROI standard gains adoption.
- A large enterprise publishes finance-controlled, GL-reconciled AI ROI with a real holdout.