Cisco just delivered a quarter that, on paper, looked pretty solid. Revenue hit $15.35 billion (beating the $15.12 billion estimate), earnings came in at $1.04 adjusted (above the $1.02 consensus), and sales grew 10% year-over-year. Product orders surged 18% across every geography.
So naturally, the stock dropped nearly 10%.
Welcome to Wall Street, where beating expectations isn’t enough — you have to beat expectations about the future, too. And Cisco’s guidance was the problem. The company projected Q3 earnings of $1.02 to $1.04 per share and revenue of $15.4 to $15.6 billion. That’s roughly in line with estimates. Not bad, not great. Just… fine. And “fine” at 25 times earnings doesn’t cut it.
The deeper issue is margins. CEO Chuck Robbins flagged rising memory costs — driven by insatiable demand for Nvidia’s GPUs — as a headwind squeezing profitability across the networking equipment space. Cisco has responded with price hikes and contract adjustments, but GAAP earnings for the full year are projected at just $3.00 to $3.08 per share, implying profits are essentially flat for the rest of the fiscal year.
There is a real AI story building inside Cisco. The company reported $2.1 billion in AI infrastructure orders from hyperscalers during the quarter, and it’s partnering with AMD on a major AI infrastructure project in Saudi Arabia. It also launched a new networking switch with an Nvidia chip inside. Robbins pointed to “neoclouds” — younger cloud providers building AI-first infrastructure — as a growth driver that should start ramping in the back half of the fiscal year and accelerate into 2027.
But that’s the rub. Cisco’s AI narrative is a 2027 story being priced against a 2026 reality of flat earnings and rising input costs. The $2.1 billion in AI orders sounds impressive until you realize that’s roughly one quarter’s worth of orders in a company doing $61 billion a year in revenue.
For traders watching this name, the 10% selloff might look like an overreaction — and it probably is, at least partly. But until Cisco can show that AI orders are translating into margin expansion rather than just revenue maintenance, the stock is likely stuck in no-man’s land. The growth is real. The valuation premium isn’t earned yet.