Here is the question that matters going into Thursday evening: if Arista Networks can grow revenue 58.5% in a single quarter, beat estimates by nearly $330 million, and still watch its stock fall more than 10% the next trading session, what does that say about what investors actually need to hear from Cisco?
The answer is that in a market this skeptical of AI infrastructure spending, one number carries the entire weight of Cisco’s valuation: the AI infrastructure order rate for fiscal Q1 2027. Everything else is secondary.
Why This Report Is Different
Cisco reports fiscal Q1 2027 results on Thursday, November 12. Wall Street expects earnings of $1.32 per share on revenue of $18.11 billion, figures that align with the guidance Cisco issued in August. Cisco has beaten EPS estimates in each of the past four quarters, so a headline beat would be almost expected.
What cannot be modeled in advance is the AI order line. Cisco booked $4 billion of AI infrastructure orders in fiscal Q4 2026, taking total orders for fiscal 2026 to $9.3 billion, and the company expects AI infrastructure revenue to reach $7.5 billion in fiscal 2027. That target implies roughly $1.9 billion of AI revenue per quarter. The first read on whether that pace is holding arrives Thursday.
The broader context makes the disclosure more fraught. Arista stock fell 10.8% on November 5 during a broad retreat from AI infrastructure stocks, with Arista’s peers falling 7.6% to 15.2% in the same session. Morgan Stanley downgraded Cisco from Overweight to Equal Weight on the same day, cutting its price target from $135 to $120. The firm cited a decline in valuation attractiveness following a significant rise in the stock price, with potential growth from Cisco’s AI business largely reflected in the current valuation.
What Arista Confirmed, and What It Left Open
Arista delivered Q3 2026 revenues of approximately $3.66 billion, exceeding guidance of $3.3 billion while growing 58.5%, the highest year-over-year revenue growth since its IPO. Looking ahead, Arista projected fourth-quarter revenue between $3.9 billion and $4 billion, comfortably above the roughly $3.6 billion consensus. By any measure, that is a confirmation that AI network spending is real and accelerating.
Yet Arista’s report did not resolve the question specific to Cisco: whether hyperscaler orders are flowing to both vendors or concentrating in one. Cisco’s Ethernet-based infrastructure competes directly with Arista in the data center, and Cisco’s version of the story runs through Ethernet, with the buyers being the hyperscalers whose orders drove Cisco’s AI infrastructure book from $1.3 billion in Q1 FY2026 to $4 billion in Q4. Thursday’s order disclosure is the cross-check.
The Bull Case
The underlying business is in genuinely strong shape. In Q4 FY2026, Cisco posted revenue of $17.3 billion, up 18% year over year, with GAAP EPS up 52%. Networking orders climbed 40%, Silicon One chips secured new design wins, and enterprise and sovereign customers continued upgrading their infrastructure. That breadth matters: Cisco is not a pure hyperscaler play, which gives it some insulation if any single customer pulls back. FY2026 AI orders of $9.3 billion were roughly 4.5 times the previous year’s total.
The stock pays an annual dividend of $1.68, yielding approximately 1.8% at recent prices, with roughly 14 consecutive years of dividend growth. For a company whose valuation now turns on AI order momentum, that income floor is not irrelevant. It draws a category of buyer that pure-growth networking names like Arista simply cannot.
The Bear Case
Cisco’s current P/E ratio is around 28.6x, meaning the stock has re-rated significantly on the expectation that AI orders sustain. If Thursday’s order number disappoints, or the Q2 outlook implies sequential deceleration, that multiple contracts fast. Non-GAAP gross margin already compressed to 66.3% from 68.4% as AI hardware carries lower margins than software. More AI revenue solves the top-line question. It does not necessarily solve the margin question.
What to Watch
Three things decide how the stock opens Friday. First, the Q1 AI infrastructure order figure, against the roughly $2 billion quarterly pace implied by FY2027 targets. Second, whether management signals that the Q2 order pipeline is building or flattening. Third, any commentary on gross margin trajectory as the mix shifts further toward hardware.
Arista proved the spending is real. Cisco has to prove it is capturing enough of it to justify a stock that has already priced in the supercycle. Thursday is when that proof arrives, or does not.
