Findings from Deloitte’s 2026 State of AI in the Enterprise report are making their way into a fair few business cases right now.
Three findings, among those surveyed, are particularly interesting when taken together:
- 66% of organisations said they saw efficiency and productivity grains from AI
- 20% said they saw revenue increases
- 74% hoped to grow revenue through AI initiatives in the future
Details matter, but for the purposes of this article, I’ll take those figures at face value.
I have two takeaways:
- If you’re writing a business case and quoting the report, your strongest groundings are productivity and efficiency
- Revenue is more of an ambition than a reality right now
TL;DR: a case can be made for efficiency, but this doesn’t automatically translate to increased revenue.
For example, if AI gives someone back five hours a week, you’ve just created capacity. Whether that capacity reduces cost, improves service or generates revenue depends on what that person uses it for.
Bridging the two
An important question for our next evolution of AI rollouts is: when efficiencies are found, can the capacity they release be used to drive revenue?
In other words, are you backing efficiency use cases with revenue opportunities? And are teams being deliberate about how they use the capacity they recover?
An example
Take a solution that reduces the time a sales team spends creating proposals. You’ve created just capacity.
But if proposal preparation was limiting how many opportunities the team could pursue, the larger benefit may come from using that capacity to have more customer conversations, build more pipeline and close more work.
Of course, this assumes a few things: there’s sufficient demand, the right incentives are in place and there’s a clear plan for redeploying saved time.
Where do we go from here
These are the questions we need to ask going forward:
- What capacity will this use case release?
- Where will that capacity go?
- What business constraint could it help remove?
- How will we know whether it changed the result?
Our opportunity ahead as an industry isn’t just to make work more efficient. It’s to know what to do with the capacity that comes from it.
About the research
Deloitte surveyed 3,235 senior leaders across 24 countries between August and September 2025. It describes the sample as organisations on the leading edge of AI, and all participating organisations had working AI implementations in daily use.
Limitations: These percentages reflect benefits reported by respondents. They do not tell us the size of those benefits, the return after implementation and operating costs, or whether reported productivity gains were ultimately converted into financial value.