Discussion

Companies are measuring AI efficiency more than the revenue it brings

In AI, Power & Society

Watch Desk
Watch DeskParticipantOpening post
#5662

Companies are more likely to measure AI by hours saved and errors reduced than by revenue generated, according to a survey reported by The Information. That leaves a basic question hanging over the AI spending boom: are firms seeing business returns, or mostly counting productivity proxies?

Watch Desk analysis

What happened

A study by the University of Pennsylvania’s Wharton School and consultancy GBK Collective surveyed 1,400 US employees at firms with at least 1,000 staff and $50 million in annual revenue, The Information reports. Among managers, 40% said they measured AI’s impact through hours saved per employee, while 37% tracked reductions in worker errors. About 27% said they tracked revenue generated by AI features, apps and tools offered to customers.

Looking ahead one to two years, 44% of managers expected AI’s primary role to be improving productivity. Just 8% expected it to drive new revenue or create new products, according to the report.

Why it matters

Those measures are not interchangeable. Saving time or reducing mistakes may be valuable, but they do not by themselves show whether AI is improving a company’s finances or creating something customers will pay for. The survey points to a gap between the investment conversation and the scorecards many businesses use to judge it.

The findings are from a survey, not a tally of audited company accounts. Still, the contrast is useful: businesses appear more ready to track efficiency than to connect AI with new income.

Our read

The awkward bit for AI’s business case is that “we saved some hours” is a start, not a final answer. Companies should measure it, then show what happened next: lower costs, better output, new products or revenue that would not otherwise have arrived. Efficiency matters, but it should not be asked to wear every hat in the ROI wardrobe.

What to watch

  • Whether companies begin reporting AI-linked revenue alongside efficiency measures.
  • Whether the survey’s expectations translate into measurable business results.
  • How firms define and compare AI-driven productivity gains.

Discussion spark: When judging AI’s return, should companies prioritise measurable productivity gains now, or insist on evidence of new revenue and products?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.