Is AI even worth it for your business? 5 expert tips to help prove ROI (www.zdnet.com)

A ZDNET panel at Informatica’s World Tour revealed that more than 97% of organizations struggle to prove generative AI’s business value, and offered five practical ways to fix that: start small and know when to stop; win hearts-and-minds across the organization; make ROI a two‑way conversation with finance; tie use cases to broader business outcomes; and tightly track project moving parts. Speakers from Jotun, Accenture, Rabobank, AWS and EDF stressed that measuring AI ROI is less about perfect math and more about pragmatic signals, stakeholder alignment, and storytelling that maps technical work to specific business metrics (e.g., churn reduction, market expansion). Technically this means investing in data foundations—cloud migrations and centralized data hubs (examples: Informatica + Snowflake stacks, with Power BI for operational dashboards)—so teams can iterate rapidly, scale winners, or kill losers. It also requires explicit governance: clear roles, targets, costs, timelines and finance collaboration to build credible business cases. For the AI/ML community the takeaway is concrete: prioritize robust data infrastructure and cross-functional metrics early, present outcomes in the language of each stakeholder, and use disciplined project controls to turn experimental models into measurable, investable value.