Three reports today, and one of them is going to make consulting analysts rethink every AI-spending assumption they had walking into this quarter.
Three reports today, and one of them is going to make consulting analysts rethink every AI-spending assumption they had walking into this quarter. Let us start with Accenture, because an eighteen percent gap down on a print that showed seven point three percent revenue growth and a ten point six…
Transcript
Three reports today, and one of them is going to make consulting analysts rethink every AI-spending assumption they had walking into this quarter.
Let us start with Accenture, because an eighteen percent gap down on a print that showed seven point three percent revenue growth and a ten point six percent net margin is not a math story — it is a guidance story. EPS grew less than one percent year over year on a base the Street had priced for acceleration, and what the market is telling you this morning is that the premium multiple for an AI-transformation beneficiary just got stress-tested. The reason it happened is specific: bookings momentum slowed enough that management could not defend the forward narrative, and when a company trading on future consulting wallet-share loses the credibility of its own pipeline math, the multiple has nowhere to go but down fast. For a services business where the entire bull case is that enterprises will spend more, not less, on large-scale technology programs, a bookings softness signal is not a one-quarter noise event — it is a leading indicator, and the Street is only starting to price what it means for the next two or three reporting cycles. The ten point six percent net margin sounds clean until you remember what Accenture is selling: people, process, and the promise of transformation. When the pipeline thins, that margin does not protect you — it just means you are losing less slowly.
Kroger next. The number I want you to hold is the price-to-earnings ratio — thirty four point four — on a business printing a zero point seven percent net margin and earnings per share down fifty seven point eight percent year over year. That is not a rounding error. That is a grocery chain with almost no margin buffer trading at a multiple that would make a growth software founder blush. The eight and a half percent gap down is not the story; the story is what you are paying for what you are getting, and right now the answer is a lot for very little. A fifty-seven percent earnings decline against flat revenue growth of zero point four percent tells you exactly where the pressure is living — somewhere inside the cost structure, and probably more than one place at once. The multiple only makes sense if you believe the earnings base is temporarily distorted. That is a bet worth examining very carefully before you make it.
Jabil is the quieter one on a day this loud, down less than one percent on the print despite a headline that actually looks constructive — nineteen percent revenue growth, earnings per share up seventy eight point eight percent year over year. So why is the market shrugging? Because the tension I see in every contract manufacturer running on two percent net margin against a price-to-earnings multiple in the mid-forties is the same tension here: the operational leverage is real, but the margin for error is not. One customer concentration shift, one component cost cycle, one end-market that softens faster than the order book adjusts — and that earnings growth story reverses faster than the revenue line can catch it. The beat was real. The question is whether the structure underneath it is as durable as the headline implies. At this margin profile, you do not get to be wrong twice.
Over the coming days I am going back into Accenture for the full teardown — the bookings disclosure, the segment mix, and the exact language management used to reframe the AI story. Kroger also deserves a closer look at how a fifty-seven percent earnings drop coexists with a thirty-four multiple, because there is a structural accounting conversation buried in that gap, and I want to find where the distortion actually lives.
That is the teardown. The numbers were always there — most people just do not look. See you at the next filing.