Sterling Infrastructure: Why The Growth Story Isn't Over (Earnings Preview)

Sterling Infrastructure reported Q4 2025 sales up 51.5% year over year and EBITDA up 70%, supported by strong data center and semiconductor construction demand. The company ended with a $3.01 billion backlog and $100 million net cash, while management guided FY2026 revenue to $3.05 billion-$3.20 billion and non-GAAP EPS to $13.45-$14.05. The article argues consensus estimates are too conservative, reinforcing a constructive outlook.

Analysis

STRL is in the sweet spot where backlog visibility converts into multiple expansion before it converts into cash flow. The market is likely still underestimating the duration of the data-center / semiconductor capex cycle because those end markets are being funded by hyperscalers and domestic industrial policy, not just cyclical GDP spending; that makes the revenue stream feel more like a secular buildout than a normal infrastructure book. The key second-order effect is that subcontractors, specialty electrical outfits, and regional site-work peers should see tighter labor and materials availability, which can widen STRL’s execution moat if it has pricing power and project discipline.

The main risk is not demand, but schedule slippage and margin normalization if a few large jobs shift right or if procurement bottlenecks hit equipment, power, or labor. Over a 3-6 month window, the stock can still rerate on estimate revisions alone; over 12-18 months, the debate becomes whether the current growth rate is repeatable once the easiest backlog burns off. If investor positioning is already crowded, any sign of margin plateauing could compress the multiple even with still-strong fundamentals.

Consensus appears too anchored to a mean-reversion model that treats this as a standard contractor rather than a beneficiary of a multi-year capacity expansion. What the market may be missing is that the backlog is not just a revenue buffer — it is a call option on accelerated data-center and AI-related infrastructure spend, which can force upward revisions across 2026-2027 if project starts stay hot. That said, the setup is more attractive on dips than on strength, because upside is driven by estimates catching up while downside is driven by any hint that the backlog is less fungible than it looks.

Key Decisions for Investors

• Long STRL on any 3-5% pullback over the next 1-3 weeks; target a 15-20% move over 3-6 months as FY2026 estimates reprice higher.
• Use call spreads rather than stock for event-driven upside: buy STRL 6-9 month call spreads struck around 5-10% out of the money to capture estimate revisions with defined downside.
• Pair trade: long STRL / short a lower-quality general contractor or infrastructure name with weaker backlog conversion and lower net cash, for 6-12 months, to isolate secular data-center exposure from cyclicality.
• Watch for any commentary on labor, electrical equipment, or power-interconnect lead times; if those worsen, trim 25-30% of the position because margin risk can emerge before revenue risk.
• If the stock gaps up sharply on guidance revisions, take partial profits into strength rather than chasing — the better risk/reward is before consensus fully resets.