⚠ Customer Concentration — the TitansHigh threat

Arista Networks (ANET) — threat to the moat

Microsoft 26% plus Meta 16% — nearly half the company rides on two buyers with all the leverage.

The single sharpest risk in Arista is its extreme customer concentration: Microsoft accounted for roughly 26% of 2025 revenue and Meta another 16%1 — about 42% of the entire business from just two customers — with the cloud and AI titans together near 48%. Concentration of this degree means Arista's fortunes are tied directly to the decisions of a few enormous, powerful buyers whose behavior it cannot control, and it is the clearest single reason the moat is rated narrow. A material change at either giant — a slowdown in spending, a shift of volume to a competitor, a move to build their own networking, or harder negotiation on price — would strike Arista's revenue and margins directly, and the two are large enough that no quick diversification could offset the loss.

Largest single customer's share of revenue (%)21%202320%202426%2025Arista Form 10-K FY2025
One customer bought more than a quarter of everything Arista sold in 2025.

The danger is compounded by who these customers are and how they behave. Microsoft and Meta are among the most sophisticated technology companies on earth, with the resources and the motive to preserve their leverage: they multi-source, they negotiate aggressively (their discounts are already compressing Arista's gross margin), they can build white-box networking or adopt open software, and they could shift toward Nvidia's networking as the AI battle unfolds. Their spending is also tied to the AI capital-expenditure cycle — enormous, but untested by a downturn — so an AI digestion phase would hit Arista's two most important customers at once. The concentration thus stacks several risks on the same narrow base: customer power, the AI cycle, the competitive battle, and the white-box threat all bear most heavily on the handful of accounts that drive nearly half of revenue.

The mitigating truths are real but partial. Arista serves these giants superbly, its deep integration and software make it hard to displace mid-cycle, the AI build-out is driving their spending sharply higher, and the company is working hard to diversify into the enterprise and campus markets to broaden its base — genuine efforts that are gradually reducing the concentration. And a customer spending billions on Arista's networking is not lightly going to rip it out. But an investor must weigh honestly that nearly half of Arista's revenue, and much of its growth, depends on a handful of powerful giants who hold the leverage, whose spending rides an untested cycle, and any one of whom could — by a strategic decision entirely within its own control — materially damage Arista's business. The concentration is the risk that dwarfs the others, the hardest to hedge, and the fundamental reason a business this excellent carries a narrow rather than a wide moat.

The number that tests this threat
Revenue from the two biggest customers
Microsoft ~26% + Meta ~16% = ~42% of FY2025 revenue (titans ~48%)

The sharpest risk, quantified: roughly 42% of revenue from just two customers, ~48% from the cloud/AI titans. They negotiate hard (already compressing gross margin), multi-source, could insource or shift to Nvidia, and their AI spending rides an untested cycle. Watch the top-customer share and any sign the giants slow or diversify away.

Source: Arista 10-K (customer concentration)
References
  1. ReportedMicrosoft ~26% and Meta ~16% of 2025 revenue — ~42% from two customers.
    Arista Networks Form 10-K, fiscal 2025 — revenue $9.01B (+29%), net income $3.51B, diluted EPS $2.75, gross margin ~64%; customer concentration disclosed (Microsoft ~26%, Meta ~16% of revenue) — FY2025 · publ. February 2026 · source ↗
Sources
Generated September 23, 2026