Garrett Fritz: How do leaders measure real AI ROI?

Garrett Fritz, Partner and CTO at MetaCTO, is a longtime CTO and product leader, and we spoke about why many companies feel intense pressure to “use AI” yet struggle to show real returns. With an aerospace engineering degree from MIT and years as an in-house CTO and head of product, Garrett has built and shipped products across media, sports, and technology before moving into fractional CTO work for startups and mid-market firms.The turning point he described was noticing a widening gap between expectations and results. Executives approve AI budgets and ask teams how they’re using it, but the honest answer is often, “we’re trying ChatGPT.” Garrett calls this pattern “AI theater,” where mandates cascade downward as “figure it out,” leaving teams with scattered tools, no shared standards, and no way to measure impact. Companies can spend “thousands, tens of thousands of dollars a month” on tools without seeing increased output or reduced costs.His approach is practical and top-down: leadership must choose which tools the company will use, define how they should be used, and then measure adoption and outcomes. This doesn’t require AI experts, but it does require ownership, time, and clear authority. Without that, even strong individual contributors hesitate to share gains because “if I let everyone know that my job is so much easier, I’m going to get fired.” The real work, Garrett argues, is aligning tools with workflows, deciding when to build versus buy, and tying usage back to delivery, reporting, and ROI.Listeners will come away with a concrete way to move past AI hype toward measurable, organization-wide results.Key takeawaysAI pressure often leads to unmeasured, scattered tool adoption“AI theater” replaces strategy when leaders say “figure it out”Pick standard tools and define how they must be usedMeasure adoption only after leadership sets clear expectationsHigh AI spend doesn’t guarantee higher output without integration