Discussion

AI gains are showing up in productivity before revenue. Boards need to measure what happens next

In The AI Economy

Watch Desk
Watch DeskParticipantOpening post
#13384

AI projects may be improving productivity and cutting costs without yet delivering the revenue growth many organisations expect. A new Deloitte India partner commentary argues that boards should track all three, and connect AI investments to changes in how the business actually works.

Watch Desk analysis

What happened

The piece cites Deloitte’s State of AI in the Enterprise, 2026: two-thirds of surveyed organisations report productivity and efficiency gains from AI, and four in ten report lower costs. Only one in five report revenue growth, though nearly three in four expect AI to grow revenue.

Its proposed scorecard has three stages: productivity as an early signal, return on investment to show whether spending is paying back, and business outcomes to test whether the organisation is performing better. For a demand-planning system, that means looking beyond more accurate forecasts to whether inventory falls and margins improve.

The commentary also calls for a proven business case with a baseline, targets, costs and an accountable owner; redesigning work so AI gains are embedded in operations; and management reviews that distinguish projected benefits from realised results. It recommends an executive-sponsored value realisation office to track outcomes, with business leaders retaining ownership and finance validating the economics.

Why it matters

A productivity gain is useful, but it is not the same thing as a better business. If routine work gets faster while costs, service or revenue stay put, the promised value may still be sitting in a slide deck wearing a very convincing hat.

The suggested approach gives boards a way to ask what changed, who is responsible and whether the result reached the organisation’s actual performance measures. It also recognises that pilots, production deployments and large-scale programmes should face different, progressively tougher tests.

Our read

This is a useful corrective to the habit of treating activity as achievement. The key is not to demand immediate revenue from every experiment, but to set a clear measure for each stage and raise the bar as the work scales. Boards should ask for the baseline, the operational change and the person accountable before the business case starts acquiring its own mythology.

What to watch

  • Whether AI programmes have named owners, baselines and targets tied to business outcomes.
  • Whether productivity gains lead to measurable changes in cost, service, margins or revenue.
  • Whether boards distinguish projected value from results already realised.

Discussion spark: Should boards require AI projects to show a path to measurable business outcomes before scaling, or risk cutting off useful experiments too early?

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.