Accenture Is the Canary in the Consulting Mine

Accenture’s stock is heading toward a nine-year low, and the market is finally pricing in what anyone paying attention already knew: the business model was always fragile.

The standard consulting frame sells human hours as a proxy for expertise. Accenture charges clients for bodies — analysts, architects, program managers — assembled in delivery centers from Bangalore to Dublin. The model scales beautifully in a world where the expertise is genuinely scarce and the labor arbitrage is real. For two decades, both conditions held. Now neither does.

Everyone says AI threatens consulting at the margins — it automates the slide decks, speeds up the due diligence, writes the first draft of the requirements doc. The opposite is closer to true. AI doesn’t just accelerate the work; it exposes that a large fraction of the work was always the performance of expertise rather than expertise itself. The billable hour was never really a measure of value creation. It was a convention the whole industry agreed to maintain because the alternative — pricing on outcomes — would have required consultancies to own their recommendations. They never wanted to own them. AI forces the question.

Here’s the number that matters: Accenture generated roughly $65 billion in revenue last year. A meaningful slice of that — the IT implementation work, the systems integration, the managed services — required genuine coordination and judgment. That slice is defensible. The rest of it, the transformation theater, the strategy frameworks repackaged and re-sold, the army of junior analysts doing work that GPT-4o now does in seconds — that slice is not. The market is trying to figure out how big each bucket is, and the bear case is that the indefensible bucket is a lot larger than management admits.

The Accenture response, predictably, has been to reposition around AI. They’ve announced multibillion-dollar investment commitments in AI capabilities. They’ve launched new service lines around AI transformation. This is the right instinct dressed up as the wrong execution. Selling AI implementation services to clients who are adopting AI to reduce their need for implementation services is a loop that eventually closes. The question is not whether Accenture can build AI practices. Of course they can. The question is whether clients will pay Accenture-level margins for AI implementation when every hyperscaler — Microsoft, Google, AWS — is also selling AI transformation services at subsidized rates to lock in cloud spend. Accenture is getting squeezed from above by the platforms and from below by the models.

The deeper structural problem is that Accenture’s competitive position was never really about knowledge. It was about certified trust. Large enterprises bought Accenture not because they thought Accenture had uniquely brilliant people, but because Accenture carried the relationship liability. If the transformation failed, you could blame the consultants. That accountability-transfer function still has value — but it’s worth a lot less when the transformation is an AI deployment and the hyperscaler is the de facto implementation partner anyway. Microsoft doesn’t just sell you Copilot. Microsoft sells you the Copilot deployment playbook, the training, the support structure. They’re eating the accountability transfer margin too.

None of this means Accenture goes to zero. The company has real capabilities, real client relationships, and real institutional knowledge in verticals like defense, healthcare, and financial services where the work is genuinely complex and the regulatory environment creates moats. The federal business alone is a significant floor. But the $75 billion market cap the stock commanded at peak required believing the entire enterprise was defensible. It wasn’t then. It definitely isn’t now.

The consulting industry built its empire on information asymmetry — the client doesn’t know what best practice looks like, so they pay someone to know for them. AI commoditizes best practice. It doesn’t eliminate the need for judgment, but it makes the judgment visible and contestable in ways it never was before. You can now ask a model to critique the consultant’s framework in real time. The asymmetry is collapsing.

The stakes here go beyond one company’s stock price. Accenture is a leading indicator for a category: the high-margin human intermediary that sits between enterprise decision-makers and the underlying work. McKinsey, Deloitte, IBM Consulting — they’re all watching this closely, and they’re all running the same repositioning playbook. The playbook might buy them a few years. It won’t solve the underlying structural problem.

When your moat is renting out other people’s expertise at a markup, the worst thing that can happen is expertise becoming cheap. That’s exactly what happened.