Thesis. A scarce datacentre site-development and mission-critical electrical execution platform with strong repeat-phase visibility β the next test is whether a backlog that more than doubled converts without margin dilution from the lower-margin electrical business.
Reviewed 19 Sep 2026, after Q2 2026 results on 3 Aug. Next review after Q3 2026 earnings on 2 Nov 2026 (expected, unconfirmed).
What it does
Sterling does large-scale site development and mission-critical electrical construction for datacentres, semiconductor fabs and advanced manufacturing. It prepares the campus β earthwork, underground infrastructure, pathways β and installs the essential electrical work before racks, power equipment and cooling can operate.
It belongs in Grid Equipment & Build alongside PWR: both supply construction capacity, while ETN and the equipment vendors supply the hardware. Sterling owns an execution-capacity bottleneck around complex sites and skilled labour; it manufactures nothing.
Two backlog figures are quoted. Signed backlog is contracted work. Combined backlog adds awards not yet under contract. The second is larger and softer.
Speed has economic value here: bringing an AI campus online sooner lets the customer monetise compute sooner, which is why successful site work tends to be renewed by negotiation rather than rebid.
Bull case
- Q2 2026 revenue $1,168M, +90% YoY, with consolidated organic growth around 50%
- E-Infrastructure revenue $905M, +192% β 78% of the company. Segment operating income $210.8M
- Adjusted EPS $5.80, +116%. Adjusted EBITDA $256.7M, +104%
- Signed backlog $4.33B (+116%); combined backlog $5.62B (+150%). E-Infrastructure backlog +165%, of which 92% is mission-critical work across datacentres, manufacturing and semiconductor facilities
- First-half operating cash flow $328M, +93%, on $70M of capex. Net cash position at 30 June
- FY26 guidance raised: revenue $4.00β4.15B, adjusted EPS $19.70β20.30, adjusted EBITDA $891β916M
- Repeat phases improve visibility β management says successful site work often leads to negotiated follow-on phases rather than a full rebid
- Vertical integration and scale (specialised equipment, 3D modelling, LiDAR, underground coordination, self-performed work) reduce schedule friction
- The CEC acquisition widens the capability set into electrical construction, increasing Sterling's share of each campus
Bear case
- Margin mix is moving against the headline growth. Legacy site-development margins ran in the high 20s; CEC's Q2 adjusted margin was ~12%. Reported E-Infrastructure margin fell to 23.3% from 27.0% purely on mix
- Management warned Q3 backlog could decline sequentially on award timing and strong revenue burn. Not automatically a demand signal, but it means organic book-to-burn matters more than the reported total
- Labour is already the binding constraint. Management said CEC could grow faster with 1,000β2,000 additional electricians
- Datacentre exposure is not separately disclosed. 78% E-Infrastructure and 92% mission-critical are not pure AI percentages β semiconductor and manufacturing work is inside both
- No proprietary technology. The moat is execution capacity, labour, equipment scale and relationships, not patented hardware
- Segment-level customer concentration: the top four E-Infrastructure customers were 27% of segment revenue in 2025 (down from 40% in 2023). No customer exceeded 10% of consolidated revenue
- Construction execution risk on fixed-unit-price and lump-sum work; acquisitions (CEC, Stone Ridge) make reported growth less comparable
- Valuation is rich and requires sustained organic awards, margin improvement and clean cash conversion
Major customers
Customer identities are not disclosed. The disclosed framework:
- Hyperscale and colocation datacentre developers requiring large-site preparation and mission-critical electrical work
- Semiconductor and advanced-manufacturing operators
- Distribution and e-commerce facilities
- Public transportation agencies and municipalities, through the lower-margin Transportation segment
- Residential and commercial builders, through Building Solutions
Top four E-Infrastructure customers: 27% of segment revenue in 2025, down from 40% in 2023. No individual customer above 10% of consolidated revenue. Concentration is meaningful at segment level even though consolidated concentration is low.
Read-through. Hyperscaler capex and new-campus announcements are the main demand input, but Sterling's revenue depends on actual land release, power availability, permits and construction sequencing β so it lags announcements by more than the equipment names do. Running the other way, site and electrical progress is an early physical-build indicator for VRT, MOD and the other facility-equipment vendors, whose gear is installed into what Sterling has built. It sits beside PWR and MTZ as construction capacity rather than hardware, and labour scarcity is both its moat and its ceiling β the same electrician shortage that supports pricing caps how fast it can grow.
What would change the view
- Organic E-Infrastructure book-to-burn β the cleanest demand test as acquisition comparisons normalise
- CEC margin progression β evidence electrical profitability can move from ~12% toward the mid-teens
- Same-business E-Infrastructure margin, which separates operating execution from acquisition mix
- Repeat campus phases converting into signed awards without competitive rebids
- Electrician hiring and retention β the immediate capacity constraint
- Free cash flow conversion after the higher $130β140M capex plan
- Any multi-quarter award slowdown caused by power, interconnection, permitting or customer financing
Update log
19 Sep 2026 β Converted to the standard template. Content was already current through Q2 and post-Q2 estimates; no figures changed. Zacks consensus estimates removed from the body β they live in the database properties, and repeating them here was the only place on the page where a stale number could contradict the columns.
17 Sep 2026 β Page rewritten post-Q2 with the margin-mix issue elevated above the headline growth, and the Grid Equipment & Build category fit stated explicitly.
3 Aug 2026 β Q2 2026 reported. Revenue $1,168M (+90%), E-Infrastructure $905M (+192%), adjusted EPS $5.80 (+116%), adjusted EBITDA $256.7M (+104%). Signed backlog $4.33B (+116%), combined $5.62B (+150%). FY26 guidance raised to $4.00β4.15B revenue and $19.70β20.30 adjusted EPS.
Probabilities: strengthen ~65% / weaken ~25% / broadly unchanged ~10%.
Sources
Research and education only β not investment advice.