Businesses may get more from AI by redesigning workflows around it, rather than simply adding it to existing tasks. A McKinsey survey cited in a Raconteur article found that leaders were 5.3 times more likely to report enterprise value when workflows had been redesigned.
Watch Desk analysis
What happened
The article describes a progression from using AI for individual tasks to improving workflows and, ultimately, rethinking work around the technology. It says only 11% of surveyed leaders considered their organisations to have reached that final stage, but 48% of that group reported realising enterprise value, compared with 13% of organisations at the first stage.
Sami Ghoche, Zendesk’s vice-president of applied AI, argues that companies should start with the business problem, not with the AI feature they want to bolt on. In customer service, for example, AI might move from drafting replies to spotting onboarding problems early and helping prevent some support requests at their source. The article also argues that AI systems need access to business tools to take actions such as changing an order, which calls for business and technical teams to work together.
Why it matters
These figures point to a workplace and organisational challenge, not simply a model-selection problem. If a company keeps the old workflow intact, it may get a faster version of the same process. Redesigning the work could open up new uses, but it also means changing who is involved, how systems connect and what staff are expected to do.
The survey figures are reported in the article, which does not provide enough detail here to assess the survey design or establish that redesign alone caused the reported difference. The distinction matters: an association is useful evidence to explore, not a guaranteed return on investment.
Our read
The strongest idea is to ask what outcome needs improving before deciding where AI belongs. That sounds less glamorous than announcing an enterprise-wide transformation, but it is a better starting point than sprinkling a chatbot over a creaking process and calling it strategy. Managers should involve the people who do the work and the teams responsible for the systems, then measure whether the redesigned process actually helps.
What to watch
- Whether further survey detail explains how the reported enterprise value was measured.
- Whether companies redesign workflows with staff input, or mainly use AI to increase output targets.
- Whether teams give AI the system access needed to act, alongside clear accountability and safeguards.
Discussion spark: If AI saves time on a task, should employers use that time to redesign the work, or expect staff to do more of the same?
Sources and evidence
- Source update (12 October 2026, 16:35 UTC)
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.