Celestica: A Bet On AI CapEx Growth

Celestica beat Q1 estimates and raised FY 2026 revenue guidance, but shares still fell 14% after earnings. The Data Center-heavy CCS segment generates 80% of revenue and grew 76% year over year, underscoring strong AI and hyperscaler-driven demand. Despite the selloff, the operating backdrop remains positive with management pointing to continued cloud hardware strength.

Analysis

The post-earnings selloff looks more like positioning unwind than a thesis break. When a company with concentrated AI-exposed revenue raises guidance and still gets marked down, that usually signals the market had already crowded into the “beat-and-raise” outcome and is now de-rating the quality of the upside. The important second-order effect is that hyperscaler hardware demand is still flowing, but investors are starting to question how much of the near-term growth is already committed vs. incremental, which can pressure multiple expansion across the AI hardware supply chain.

The competitive implication is favorable for diversified contract manufacturers and subsystem suppliers that can absorb share if one node stumbles or gets capacity-constrained. If Celestica keeps winning AI infrastructure work, peers with weaker exposure to AI clusters may still benefit from reallocation of OEM spend and tighter lead times elsewhere in the chain. The real risk is not demand disappearance but customer concentration: any single hyperscaler capex pause can create sudden order air pockets, and those are typically priced 1-2 quarters ahead rather than at the point of reported softness.

On timing, the stock likely trades more on sentiment and incremental order commentary over the next few weeks than on the raised guidance itself. If AI capex checks stay firm, the drawdown can reverse quickly; if not, the market will treat this as an early warning that the AI hardware cohort is moving from scarcity premium to execution premium. The contrarian takeaway is that the selloff may actually improve the risk/reward for investors willing to own the name through a volatile digestion phase, because the market is now demanding proof rather than paying for narrative.

I would frame this as a tactical long rather than a blind momentum buy: the setup is attractive if the next order/tape commentary confirms hyperscaler demand remains intact, but it is vulnerable to any broader AI hardware multiple compression. In other words, this is a stock where fundamentals may still be improving even as the valuation regime tightens.

Key Decisions for Investors

• Buy CLS on weakness over the next 3-5 trading days with a tight stop below the post-earnings low; target a 10-15% rebound if AI-capex sentiment stabilizes and the market re-prices the guidance raise.
• Use a call spread in CLS with 1-3 month tenor to express a rebound in sentiment while limiting downside if the market continues to de-rate AI hardware names.
• Pair long CLS / short a lower-quality or less AI-levered hardware peer over 4-8 weeks to isolate execution alpha from sector beta; thesis is that confirmed AI infrastructure demand should reward the cleaner operator.
• If you are already long the AI hardware basket, hedge with a small CLS short against the basket for 1-2 weeks only if you expect follow-through selling from crowded positioning; cover quickly if order commentary improves.
• Set a catalyst watch for next channel checks or customer commentary over the next 30-60 days; if hyperscaler demand is reaffirmed, add aggressively because the current move looks more like a multiple reset than a fundamental inflection.