Cisco (CSCO) Down 5.3% Since Last Earnings Report: Can It Rebound?

Cisco's fiscal Q4 2026 non-GAAP EPS rose 23% year over year to $1.22, beating consensus by 4.27%, while revenue increased 18% to $17.25 billion and exceeded estimates by 2.36%. Networking revenue grew 28% to $9.79 billion, supported by $4.0 billion of hyperscaler AI infrastructure orders in the quarter; Cisco expects hyperscaler AI revenue to rise to $7.5 billion in fiscal 2027 from about $4.0 billion. Fiscal 2027 guidance calls for $72.2-$73.4 billion in revenue and $5.05-$5.11 in non-GAAP EPS, although gross margin declined 210bps to 66.3% amid a higher hardware mix and memory costs. Shares are down 5.3% since earnings, but consensus estimates have risen 19.25% over the past month and the company holds a Zacks Rank #2 (Buy).

Analysis

The relevant debate is not whether Cisco can participate in AI spending, but whether its incremental AI revenue earns a networking multiple or remains a lower-margin, customer-concentrated systems business. Hardware mix and memory costs are already diluting gross margin; if component inflation persists while hyperscalers retain purchasing leverage, operating leverage can stall despite strong revenue. This makes the next two quarterly gross-margin guides more consequential for valuation than another order beat.

Cisco's design-win momentum creates a credible 6-18 month challenge to Arista Networks (ANET), particularly in scale-out Ethernet and optics, but the near-term read-through is mixed: higher Cisco AI deployment may support component vendors such as Broadcom (AVGO), Coherent (COHR) and Lumentum (LITE), while shifting share rather than expanding total hyperscaler capex. The more durable upside is enterprise networking and security cross-sell: successful conversion of installed networking customers into Splunk-led observability and security subscriptions would raise recurring mix and justify multiple expansion. That conversion is not yet demonstrated by aggregate order figures.

The market's apparent discount likely reflects skepticism that AI orders convert cleanly into revenue and free cash flow, rather than a lack of awareness of demand. A 1-3 month rerating requires evidence that backlog conversion is not being offset by pricing concessions or memory inflation; a break below management's gross-margin floor, or a material slowdown in non-hyperscaler orders, would falsify the rebound thesis. Conversely, a sustained security/software acceleration would make Cisco less cyclical and reduce downside in an AI-capex digestion scenario.

Key Decisions for Investors

• Initiate a 1-3 month tactical long CSCO versus short ANET in equal-dollar terms only if CSCO holds its post-earnings low; thesis is incremental Ethernet share recognition, with a 8-12% relative-return target and exit if CSCO's next gross-margin guide falls below 65% or ANET reports reaccelerating cloud revenue.
• Maintain long AVGO as the cleaner picks-and-shovels exposure to networking build-outs; use any CSCO-driven demand validation to add, but monitor custom-silicon commentary and hyperscaler capex plans because Cisco share gains do not necessarily increase industry spend.
• Do not underwrite a standalone CSCO multiple expansion until disclosure shows software/security growth materially outpacing hardware and recurring revenue growth accelerating; treat this as a watch item for the next earnings release rather than a 6-18 month core long.
• For downside hedging around the next report, favor a CSCO put spread rather than outright puts if implied volatility is elevated: the principal downside catalyst is margin-guide disappointment, while substantial buybacks and yield should limit a broad de-rating absent weaker enterprise demand.