What the Best CIOs Are Telling Their Boards About AI ROI
Boards want to know what two years of AI spending has produced. The strongest answers are rarely a single number.
Boards have approved AI budgets on faith for the past two years. That patience is running out. Directors now want to know what the money has produced, and CIOs who have handled those conversations well say the answer is rarely a single return-on-investment figure.
Start with outcomes, not models
The strongest board updates avoid technical detail almost entirely. Instead of describing models, platforms and pilots, they report changes in business outcomes: faster claims processing, shorter sales cycles, fewer customer complaints, lower cost per transaction.
“The board doesn't want to hear about our AI. They want to hear about our business, and how AI changed it.”
Be honest about the portfolio
CIOs describe presenting AI as an investment portfolio rather than a list of successes.
- A few use cases delivering measurable value at scale
- A larger group still being proven, with clear decision points
- Experiments that were stopped, and what was learned from them
Admitting what did not work tends to increase the board's confidence in what did.
Separate productivity from profit
Many AI gains show up as time saved rather than money saved. The best updates explain how that capacity is being used — redeployed to higher-value work, absorbing growth without new hiring, or improving service quality — so directors can see the link to financial performance.
Prepare for the risk questions
Every AI update now attracts questions about risk: data protection, regulation, reputational exposure and dependence on a small number of vendors. CIOs who bring a clear governance summary alongside their results spend far less time defending the programme.