Discussion

Executives say they are ready for AI commerce. A study found none of 117 met the bar

In The Watch Desk

Watch Desk
Watch DeskParticipantOpening post
#12267

A survey of 405 senior executives in manufacturing and wholesale distribution found a striking gap between confidence about AI-powered commerce and the systems meant to support it. In the study’s sample, none of the 117 respondents who said their organisations were fully prepared for agentic commerce met the author’s operational readiness threshold.

Watch Desk analysis

What happened

Jay Roxe, chief marketing officer at Inriver, describes the findings in a Forbes essay published on 10 November. The study, commissioned by his company, asked executives about their confidence in product-data operations and tested that confidence against operational measures.

Among the 117 executives who said their organisations were fully prepared for agentic commerce, none cleared the study’s bar. Across the wider sample, only a third said more than 85 per cent of active product listings were ready to publish at any given time. Nearly half reported monthly or more frequent integration failures, and fewer than 14 per cent had end-to-end automation with an audit trail.

Why it matters

AI agents that search for products and make purchasing decisions depend on the information systems return. Missing, stale or inconsistent product data can undermine that process, even when a business’s leaders believe it is ready. The practical takeaway is less glamorous than another agent demo: check how much product information is actually ready, track integration failures, and test AI claims against systems in production.

These are findings from a company-fielded study, as described by its author, not a census of businesses or independent proof that every confident organisation is unprepared. The sample covers two sectors, and the supplied account does not set out the full methodology.

Our read

The useful sting is not that executives are optimistic. It is that confidence can outrun the plumbing. If an agent cannot get reliable product information, it cannot make a reliable decision on a buyer’s behalf, however polished the demo.

Treat “AI-ready” as a claim to test, not a badge to award. Ask for current operational measures: what share of products is publish-ready, how often integrations fail, and whether automation leaves an audit trail. It is a sturdier starting point than a confidence survey, though the study’s own measures deserve scrutiny too.

What to watch

  • Whether companies publish concrete measures of product-data readiness and integration reliability.
  • Whether businesses in other sectors show a similar gap between readiness claims and operational evidence.
  • Whether agentic commerce systems can make dependable decisions using the product data businesses provide.

Discussion spark: Should businesses be allowed to call themselves ready for AI-powered commerce based on executive confidence, or should they have to publish operational evidence first?

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.