Accenture's June 18 earnings report gave investors the number they didn't want: revenue missed, bookings slipped, full-year growth guidance narrowed to 3% to 4%, and the stock fell to levels last seen in 2017. This isn't a routine consulting wobble. It's a warning about how fast the old enterprise IT services model is being repriced.
If you sell large technology projects to big companies, Accenture is the name you watch because it is too large to dismiss as a one-off. The company reported $18.7 billion in revenue for its fiscal third quarter, up 6% in U.S. dollars but only 3% in local currency, while adjusted earnings of $3.80 a share came in ahead of Wall Street expectations. That split tells you the problem. Profit held up, but demand didn't look as strong as investors needed it to look.
According to Investors.com, analysts had expected about $18.8 billion in revenue, and new bookings fell 3% to $19.3 billion. The Financial Times reported that Accenture shares fell more than 15% after the report and touched their lowest level since 2017. A company with Accenture's scale doesn't usually get punished that hard for a narrow revenue miss. It gets punished that hard when investors think the story underneath the numbers has changed.
The guidance did the damage. Accenture now expects full-year revenue growth of 3% to 4% in local currency, down from its earlier 3% to 5% range. MarketWatch also noted that the company forecast fourth-quarter revenue of $17.75 billion to $18.4 billion, below the $18.48 billion analysts tracked by FactSet had modeled. That is a small numerical gap with a large psychological effect. Investors were already nervous about IT services. Accenture gave them permission to be more nervous.
The federal business is part of the squeeze, but it isn't the whole story. U.S. government spending has weakened, and Accenture has been one of the clearer examples of how Washington's pullback can hit contractors that once looked insulated. The Financial Times reported that chief executive Julie Sweet also pointed to slower client decision-making and a greater-than-expected $100 million revenue hit tied to conflict in the Middle East.
Those are real headwinds. Frankly, they still don't fully explain the selloff.
AI Is Now a Revenue Question
The harder issue is whether artificial intelligence is starting to eat the work that firms like Accenture used to bill for by the hour. You don't need to believe AI will replace consultants tomorrow to see the pressure. If clients can automate code migration, testing, documentation, customer support workflows and some low-end process analysis, they have less reason to buy the same old project teams at the same old price.
Accenture knows this, which is why it has spent the past two years trying to move the conversation from labor-heavy consulting to AI transformation. The company reported $2.2 billion in AI-related bookings in its fiscal first quarter, nearly double the year-earlier figure, according to Barron's coverage at the time. But the latest quarter brought a different signal: total bookings slipped, and Investors.com noted that Accenture has said it will no longer break out artificial intelligence bookings separately.
That change may be harmless reporting housekeeping. It may also make investors less willing to give the company credit for an AI pipeline they can't see clearly. When the stock is falling and bookings are shrinking, opacity doesn't help you.
Accenture isn't sitting still. The Wall Street Journal reported that the company is making a roughly $4.18 billion push into industrial cybersecurity, taking a majority stake in Dragos and buying runZero and NetRise outright. Dragos, based in Hanover, Maryland, protects operational technology systems, the machinery and control networks that keep utilities, factories and other physical infrastructure running. That is a real business, not a vague AI slogan.
It also shows the bind Accenture is in. The company needs new areas where clients still need high-trust technical help, and cybersecurity around physical infrastructure is one of them. But big acquisitions create their own questions. MarketWatch reported that J.P. Morgan analyst Tien-Tsin Huang raised concerns about integration risk after the deal announcements. You can understand why. Buying your way into growth is expensive when investors are already questioning organic demand.
The Message for IT Services Is Blunt
Accenture's size is the reason this report matters beyond Accenture. The company had nearly $70 billion in fiscal 2025 revenue and serves clients across governments, banks, health systems, retailers and manufacturers. When a business like that trims guidance, you should read it as a broad signal from enterprise technology budgets, not just a bad Thursday in one stock.
Capgemini fell on the same day, as Investing.com noted in its market coverage, which tells you investors saw a sector problem. The old services model depended on complexity expanding faster than clients could manage it. Cloud migrations, ERP upgrades, compliance work, outsourced operations, application maintenance, you name it, all created recurring demand for outside help. AI doesn't erase that complexity. It changes who captures the fee.
That is the issue for private equity as well. Firms that bought software services and IT consulting assets at rich multiples assumed enterprises would keep paying outside specialists to manage sprawling systems. Some will. But if automation reduces headcount needs and clients push harder on discretionary projects, the margin assumptions behind those roll-ups get thinner. The playbook hasn't broken, but the easy math has.
Accenture's fall to 2017 share-price levels doesn't mean the company has gone backward by nine years. It means the market is resetting the price of a business that once looked like a safe way to own digital transformation. Investors no longer want the phrase. They want proof that AI creates more revenue for Accenture than it removes.
The next few quarters will answer the only question that matters. If AI work turns into larger bookings and cybersecurity acquisitions add durable growth, this selloff will look too harsh. If clients keep delaying projects while automating more of the work in-house, June 18 will look less like a bad earnings day and more like the moment the consulting model started to show its age. Don't look away.
Also read: Google DeepMind puts $10 million behind the safety problem that agentic AI has been ignoring • NextEra Energy is becoming the infrastructure layer that powers the AI economy • OVHcloud is training frontier AI models and the Anthropic export ban just made its pitch a lot easier
