connected operating model<\/a> draws on McKinsey’s finding that nearly 90% of organizations use AI in at least one function yet most report no measurable EBIT impact, Gartner’s data showing successful AI organizations invest four times more in data foundations, and RAND’s estimate that enterprise AI fails at a rate above 80%, roughly twice that of conventional software. The argument: ERP, supply chain, data governance, security, and human judgment must run as one system, not five adjacent ones.<\/p>\nWhat this means for your business<\/h2>\n
Where you sit on this depends on one diagnostic question: can your organization trace a single business decision, say an invoice exception or a supplier shortage response, from the triggering data through every approval, exception path, and audit record without anyone assembling context manually? If the answer is no, you’re not behind on AI tools. You’re behind on the plumbing that makes any tool produce a defensible outcome. That gap is the actual transformation problem, and it predates your current AI budget cycle.<\/p>\n
The argument holds, and the data behind it is unusually honest for a piece published by a platform that sells into the ERP consulting ecosystem, where the incentive typically runs toward optimistic timelines and technology-first framing. Kramer inverts that tilt, insisting that governance, decision rights, and data ownership come before automation, not after. The S&P Global figure is the sharpest detail: companies abandoning most of their AI initiatives before production jumped from 17% to 42% in a single year. That’s not a maturity curve. That’s organizations discovering mid-deployment that they don’t own a decision well enough to automate it.<\/p>\n
The concept Kramer circles but doesn’t quite name is decision traceability, the ability to reconstruct, after the fact, which data an AI agent accessed, which rule it applied, which human approved the result, and what the reversal path was. Security teams think about this in terms of identity and access logs. Finance thinks about it as audit evidence. Neither group typically owns the cross-functional design that connects those two. The CIO who fills that ownership gap before the next agent deployment is the one whose program survives a compliance review or an audit when something goes wrong.<\/p>\n
The falsification condition for this whole argument is straightforward: if organizations that skip operating model redesign and just layer more capable models onto existing workflows start showing durable EBIT improvement by late 2026, the “connected model first” thesis is wrong. Right now the evidence runs the other direction, and the acceleration in abandoned projects suggests the window for avoiding expensive restarts is narrowing faster than most AI roadmaps assume.<\/p>\n
Based on reporting from Why Enterprise AI Needs a Connected Operating Model<\/a>, originally published 2026-08-05 13:19:00.<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"Share with your CIO Most enterprise AI programs are failing not because the models are wrong but because the operating model underneath them was already broken. Robert Kramer’s case for a connected operating model draws on McKinsey’s finding that nearly 90% of organizations use AI in at least one function yet most report no measurable […]<\/p>\n","protected":false},"author":1,"featured_media":7988,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[185],"tmauthors":[],"class_list":["post-7987","post","type-post","status-publish","format-standard","has-post-thumbnail","category-ai-data","tag-cio"],"_links":{"self":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/posts\/7987","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/comments?post=7987"}],"version-history":[{"count":0,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/posts\/7987\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/media\/7988"}],"wp:attachment":[{"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/media?parent=7987"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/categories?post=7987"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/tags?post=7987"},{"taxonomy":"tmauthors","embeddable":true,"href":"https:\/\/workai.tv\/news\/wp-json\/wp\/v2\/tmauthors?post=7987"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}