Top AI adopters redesign the work, not just the tools
Few CEOs see measurable AI revenue gains. The companies that do are redesigning workflows and decision rights instead of layering AI on top.
Return on AI: Weekly Digest for October 12, 2026
Forty-one findings across three themes: measuring AI returns, deployment and workflow redesign, and AI-driven cost and labor changes.
Our take
Measurement is still the weak point. Only 12% of CEOs report measurable revenue gains from AI, and more than half of enterprise leaders have no cost-per-outcome metric for the models they pay for.
The companies getting ahead are changing how work is organized, not just which tools people use. Top adopters redesign workflows entirely. Forrester finds healthcare AI stalls when workflow redesign is skipped, and Copilot deployments fail when decision rights stay the same. The labor effects are arriving too: Snap cut 1,000 jobs as AI generated 65% of its new code.
What to do about it
- Redesign the workflow and who decides what before rolling out new tools.
- Track cost per outcome for every model and agent you pay for.
- Validate AI output before it compounds into downstream errors.