MasTec

Phase A — Understand the business

Company Overview

MasTec is a North American infrastructure engineering & construction (E&C) contractor — it does not own assets that generate recurring tolls; it builds and maintains the physical infrastructure that utilities, telecoms, pipeline operators, and data-center developers need. Coral Gables, FL (Florida-incorporated), NYSE: MTZ, ~36,000 employees, 810 locations, 95+ years of operating history (incl. predecessors), top-5 in ENR's Top 400 Contractors.

How it makes money — five reportable segments (FY2025 revenue):

Contract structure & payment terms: A "significant portion" of revenue runs under master service agreements (MSAs) — generally multi-year, but crucially cancelable on short or no notice and with no minimum-purchase commitment. The remainder is project-specific. Billing terms are net-30 with 5–10% retainage held to completion; subcontracts carry "pay-if-paid" clauses that push timing risk down the chain. This is a labor-and-equipment-based, working-capital-intensive, project-margin business — not a subscription annuity. The MSA "recurring" framing is softer than it sounds: the customer can walk.

Customers/suppliers/competitors: ~1,800 customers; top-10 customers = 34% of FY25 revenue (down from 35% FY24, 38% FY23 — concentration easing). Customers are blue-chip utilities, telecom carriers (AT&T-type), midstream operators, and increasingly hyperscaler-adjacent data-center developers. Suppliers = equipment OEMs, steel/cable/pipe, fuel. Competitors: Quanta Services (PWR, the bellwether), EMCOR (EME), Primoris (PRIM), Dycom (DY, fiber-heavy), AECOM/Fluor on large civil.

The one-sentence model: MasTec rents out skilled crews + a $3.9B equipment fleet to build the wires, pipes, fiber, and power plants that the electrification/AI/energy-transition capexCapital expenditureMoney spent on long-lived things — buildings, machines, servers — rather than on running costs. super-cycle requires — and gets paid on project completion.

Supply Chain

MasTec sits in the middle of the infrastructure value chain — it is the integrator that converts capital + materials into built assets.

UPSTREAM INPUTS MASTEC END CUSTOMER / DEMAND
───────────────── ────── ──────────────────────
• Skilled labor (union + non- Engineering, procurement, • Investor-owned utilities
 union; 36k employees) construction, maintenance (Power Delivery T&D)
• Equipment fleet ($3,887M gross, across 5 segments; • Telecom carriers & towercos
 $1,728M net PP&E) — owned + 810 locations (Communications / fiber)
 finance-leased • Midstream / gas utilities
• Steel, cable, conductor, pipe, ─── chokepoints ─── (Pipeline)
 transformers, solar modules, • Skilled-labor availability • Renewable developers / IPPs
 batteries (subcontracted / (the binding constraint (CE&I generation)
 customer-furnished on large in a boom) • Hyperscalers (indirect, via
 projects) • Transformer & high-voltage data-center civil + power)
• Fuel / diesel equipment lead times • Government (IIJA/IRA-funded
• Subcontractors (pay-if-paid) • Permitting throughput broadband, water, roads)
 • Long-tenor working capital
 (DSO 65 days)

Named stakeholders from the filings: equity JVs Waha JVs (33% — Trans-Pecos & Comanche Trail gas pipelines to the Mexico border, ~$290M carrying value), Cross Country Infrastructure (CCI) (15%, a related party — see Lens 9), FM Tech (50%). Customer-side named demand drivers cited by management: AT&T / BlackRock "Gigapower" fiber JV, BEAD ($42B) / RDOF ($20B) broadband programs, LNG export terminals.

Single-source / chokepoint read: The genuine bottleneck is skilled craft labor — in an electrification boom, every E&C contractor competes for the same linemen, pipefitters, and electricians, and MasTec's own MD&A repeatedly cites "reduced productivity" and "project efficiencies" as the swing factor in margins. Secondary chokepoints: transformer/high-voltage gear lead times (an industry-wide 1–2yr backlog) and permitting. MasTec is not meaningfully exposed to a single supplier; its supply-chain risk is labor cost/availability and the timing of customer capex releases.

Competitive Advantages (moats)

MasTec's moat is real but shallow-to-moderate — it is a scale-and-relationships moat, not a structural one.

• Scale + breadth (the primary moat): One of only a handful of contractors (MasTec, Quanta, EMCOR) that can self-perform across telecom, power delivery, generation, and pipeline at national scale. The cross-segment thesis is genuine — an AI data center needs fiber (Comms), a substation + transmission tie-in (Power Delivery), new generation (CE&I), and sometimes a gas lateral (Pipeline). MasTec can bid the whole stack; a single-vertical contractor cannot. Management explicitly markets this convergence.
• Switching costs (moderate): MSAs + preferred-vendor status + safety/qualification records create stickiness, but the MSAs are cancelable, so this is a soft lock-in. Re-qualifying a new contractor on a live grid is costly and slow — that, more than contract paper, is the real switching cost.
• Equipment fleet + skilled labor pool (capital moat): A $3.9B gross fleet and 36k trained crews are hard to replicate quickly; in a labor-short boom this is a genuine barrier to a new entrant scaling up.
• Bargaining power — mixed. Over customers: weak-to-moderate (utilities and carriers are large, sophisticated, and concentrated — top-10 = 34%); MasTec is a price-taker on competitive bids but gains leverage where capacity is scarce (today). Over suppliers/subs: moderate (pay-if-paid pushes risk down; scale buys volume terms).
• What it is NOT: no network effects, no IP/patent moat, no software-style margins. Gross margins ~12.5%, segment EBITDA margins 7–15%. This is a cyclical capital-goods/services business dressed in a secular-growth narrative. The narrative is the moat's amplifier, not the moat.

Verdict on the moat: durable enough to stay a top-2 player and capture share of a rising market; not durable enough to defend pricing in a downturn. The moat is the cycle.

Segments

Segment revenue + EBITDA, FY2025 vs FY2024:

Geography: Primarily U.S. with Canada and some Mexico (Waha JVs export to the Mexican border); the filing does not break out a clean geographic revenue table, so a precise US/non-US split is n/a — not disaggregated in filing.

The Q1 2026 inflection changes the segment story:

Read: Pipeline is the surprise — a +91% revenue surge at a 21% margin on a midstream gas buildout (record 2026 US pipeline capacity additions of 18 Bcf/d cited by management). CE&I is the volume engine (data-center civil + renewables, backlog +65% YoY). The two segments the market loves (Power Delivery, Comms) are growing steadily but at thinner margins. The growth is broad-based and real; the margin story is still a work-in-progress (consolidated EBITDA margin only 6.7% in Q1 — seasonally low, but well below the 8.0% full-year).

---

Phase B — Measure performance

Earnings Result (latest print — Q1 2026, reported 2026-04-30)

A blowout-and-raise. The single most important data point in this dossier.

• Revenue $3,828.8M, +34.5% YoY (organic +29%, acquisitions +$169M). Record Q1.
• Operating income $141.8M (3.7% margin) vs $36.2M (1.3%) — +292%. Costs of revenue fell 160bps to 87.5% of revenue on better project efficiency (mostly Pipeline).
• Net income to MasTec $60.8M vs $9.9M; Adjusted diluted EPS $1.39 vs $0.51 — +173%.
• 18-month backlog: record $20,328M, up from $18,963M (Dec'25) and $15,880M (Mar'25) — +28% YoY. CE&I backlog +65% YoY.
• Guidance RAISED: FY26 revenue to $17.5B midpoint (from $17.0B), adjusted EBITDA to $1.5B midpoint, adjusted diluted EPS to $8.79 midpoint (a 33–34% raise vs prior). This is management guiding ~+22% revenue / +30% EBITDA / +34% EPS for the full year.
• Balance-sheet flags: Cash fell to $274M (from $396M) — but that's because MasTec deployed $267M on a Q1 acquisition, not operational bleed. AR rose to $3.9B (from $3.5B) on higher volume; one Q1 item: a $7.9M other-than-temporary impairment on an equity-method investment (in the Other segment) — small, non-core.
• Market reaction: stock rallied on the print; MTZ is up +119% over the trailing 52 weeks.
• Unusual vs its own history: A 34% organic-led revenue quarter and a 173% adjusted-EPS jump are far above MasTec's historical ~mid-teens growth. The Pipeline +91% / 21%-margin quarter is an outlier — partly project timing/mix, so do not annualize the 21% margin.

One-line: This was not a beat — it was a regime-change print, and management raised the year hard enough that the burden of proof has flipped to the bears.

Earnings Calls (sentiment trend)

Transcripts are not on disk (transcripts/ empty), so this is ``-grounded from release commentary and coverage.

• Tone trajectory (FY24 → Q1 FY26): apologetic-recovery → confident → emphatic. Through 2023–24 management was defending a renewables air-pocket and pipeline lumpiness (FY23 had a GAAP net loss of $49.9M ). By Q4'25 the language shifted to "new growth phase", and Q1'26 is unambiguously bullish — "record backlog," "raised guidance," "AI-powered data centers, 5G, clean energy" as the recurring triad.
• Recurring phrases management now leans on: "diversified portfolio," "convergence of trends," "data center," "electrification," "record 18-month backlog." José Mas's long-standing "$10B revenue goal" (set in 2021) is now in the rear-view at $14.3B FY25, guiding $17.5B.
• What they stopped saying: the 2023-era hedging about renewable-project delays and IRA-policy uncertainty has receded — even as OBBBA (July 2025) accelerated the clean-energy tax-credit phaseout, management reframes it as "renewables still 93% of 2025 capacity additions" and pivots emphasis to data-center-driven, policy-agnostic electricity demand. Watch this: the bullish tone now assumes the data-center demand wave replaces any IRA-credit air-pocket. That substitution is the thesis's load-bearing assumption.

Comps

Infrastructure E&C peers. Multiples are `` (stockanalysis.com via search, 2026-06) or n/a. Market caps approximate.

Read: MTZ trades at a clear premium to the pure E&C cohort (EMCOR 26x, Primoris 27x P/E) but at a discount to Quanta (48x). On EV/EBITDA the same ordering holds: PWR 40x > MTZ 26x > EME 18x > PRIM 17x. The market is paying up for MasTec's growth re-acceleration and diversified data-center exposure, ranking it second only to Quanta. The premium is defensible only if the +20–30% growth persists — at EMCOR-like mid-teens growth, MTZ would be a 25–30x P/E stock, implying material downside from $380. The valuation is the bear case (Lens 12/13).

Stock-Price Catalysts (what moves MTZ >5%)

Mostly ``; pattern inferred from filings + coverage.

• Quarterly earnings + guidance revisions are the dominant catalyst. The +119% 52-week move is overwhelmingly an earnings-and-backlog story: Q4'25 ("new growth phase," +81% GAAP NI) and Q1'26 (+34% rev, guidance raised 33–34%) were both up-moves. The 52-week high climbed from ~$291 to ~$389.
• Backlog prints move the stock nearly as much as EPS — the record $20.3B / +65% CE&I backlog was a headline driver.
• AI / data-center capex headlines (hyperscaler capex announcements, power-demand studies) are a sympathy catalyst — MTZ trades as a derivative of the AI-electricity theme alongside PWR, GEV, ETN.
• Policy shocks (down-side): the OBBBA clean-energy credit phaseout (July 2025) is the live policy risk; renewable-project cancellation headlines historically hit CE&I sentiment.
• Pipeline/commodity headlines: less so now, but a midstream capex pullback would hit the (currently high-margin) Pipeline segment.

What the pattern reveals: the market reacts to growth durability signals — backlog and guidance — more than to the headline EPS beat itself. MTZ is priced as a momentum/theme vehicle; the risk is that the same sensitivity works in reverse on the first guidance cut or backlog stall.

---

Phase C — Judge people & books

Management

• Track record (strong, founder-led). José R. Mas — CEO since 2007; Jorge Mas — Executive Chairman (brothers; the Mas family built the modern MasTec). The 2021 strategic transformation José Mas led — acquiring INTREN, Henkels & McCoy (2021), and IEA (2022, the move into union renewables + T&D) — is now paying off: revenue roughly tripled from the ~$5–8B era to $14.3B FY25, guiding $17.5B FY26. He set a "$10B revenue" target in 2021 and blew past it. That is a real, quantified, delivered transformation.
• Tenure & skin in the game (high). Jorge Mas (Chairman) + José R. Mas (CEO) beneficially own ~23% of shares outstanding as of 2025-12-31. Founder ownership at this scale aligns them with shareholders and explains the aggressive, growth-first capital allocation.
• Capital-allocation history (acquisitive, leverage-aware). MasTec is a serial acquirer — multiple bolt-ons every year (4 in 2023, 3 in 2024, 4+ in 2025, plus a $262M Q1'26 deal). Post-2021-transformation they then de-levered hard (FY24 financing: $1.15B of net debt repayment) and have kept net leverage ~1.7x EBITDA. They resumed buybacks ($77M in FY25, after none in FY24). ROE is low-quality to judge (FY23 was a loss), but the trajectory — net income +112% FY25, EPS +66% — is improving sharply. Grade: aggressive but disciplined; the M&A integration risk is real (see Lens 13).
• Founder vs professional manager: unambiguously founder-operated — the Mas family controls the board (classified/staggered, ~63% women/minority, single-class one-share-one-vote), the strategy, and ~23% of the equity. This is a feature (long-term orientation, conviction) and a governance risk (related-party density, takeover-proofing).
• Red flags — the Mas-family related-party web (material, see Lens 10/13). The most notable: MasTec performs construction services for Inter Miami CF (the MLS franchise majority-owned by Jorge & José Mas) — $77.6M of revenue in FY25 (up from $24.9M FY24, $10.7M FY23) with $37.5M receivable outstanding. Plus: aircraft leasing from a Jorge-Mas-owned entity ($5.6M/yr), equipment rentals from CCI (chaired by Juan Carlos Mas, a family member — $6.4M/yr), and $200M/$75M split-dollar life-insurance policies on the two executives. None is individually huge against $14B revenue, but the 3x ramp in Inter Miami billings and a $37.5M related-party receivable deserve scrutiny — it's company resources building the family's soccer stadium, disclosed but governance-adjacent.

Forensic Red Flags

Forensic-analyst lens. Every figure unless noted.

• Revenue recognition (the #1 thing to watch). ~Half of revenue is percentage-of-completion / over-time on fixed-price and unit-price contracts. Unbilled receivables (contract assets) = $2,001.9M, up from $1,555.8M — growing faster than revenue in absolute terms, and DSO rose to 65 days (from 60). This is the classic E&C soft spot: aggressive cost-to-complete estimates or change-order optimism inflate current revenue. Total AR (billings + unbilled + retainage) = $3.5B (Dec'25) → $3.9B (Mar'26). Not alarming yet, but the unbilled/DSO creep is the single metric to monitor each quarter.
• Cash flow vs earnings — the real flag. FY25 operating cash flow fell to $545.7M from $1,121.6M even as net income rose — a ~$576M decline driven by working-capital build (AR/DSO). FCFFree cash flowCash left after paying to run and maintain the business. Unlike profit, it is hard to flatter with accounting choices. proxy ≈ $285M (OCF $545.7M − capex $260M), a thin ~0.95% of revenue and well below net income — earnings are running ahead of cash. In a fast-growth E&C ramp this is partly mechanical (working capital funds the growth), but it is exactly where over-aggressive POC accounting would hide. Watch FY26 OCF converging back toward EBITDA.
• Segment reporting: restructured Q1 2025 (moved a utility component from Comms to Power Delivery); recast historicals provided, low manipulation risk but reduces period comparability.
• SBC & non-GAAP gap (moderate). Adjusted diluted EPS $6.55 vs GAAP diluted $5.07. The $1.48 add-back is mostly intangible amortization ($131.2M) + SBC + contingent-consideration fair-value swings — legitimate but standard "adjusted-EPS flatters reality" caution. SBC is modest (~$0.10/qtr).
• Goodwill/intangibles: carries goodwill from the 2021–22 acquisition spree; FY25 qualitative impairment test passed, no impairment. A growth stall would put this at risk (management flags it explicitly).
• Leases: large finance-lease book ($405.3M of "finance lease & other obligations" inside total debt) — appropriate for an equipment-heavy contractor, fully disclosed.
• Related parties: see the Inter Miami / aircraft / CCI web. Disclosed, audited, but a governance texture a forensic analyst flags.

Regulatory findings (required sub-section):

• SEC Litigation Releases (LR): NONE for MasTec, 2021–2026.
• SEC AAERs: NONE for MasTec, 2021–2026.
• 10-K Item 3 (Legal Proceedings): cross-references Note 15 (Commitments & Contingencies); discloses only ordinary-course litigation (personal injury, workers' comp, employment, contract/change-order disputes, FLSA wage-and-hour) with no specific material matter named. Self-insured up to deductibles via a captive; gross unpaid-claims liability $306.8M.
• Non-SEC enforcement (web search — "MasTec" (FTC OR DOJ OR FDA OR CFPB OR "consent decree" OR settlement OR fine OR penalty) enforcement): no material federal enforcement action surfaced in the search window.
• Conclusion: No material regulatory or legal findings — verified via SEC EDGAR EFTS (LR, AAER), web search, and 10-K Item 3 as of 2026-06-22. A clean accounting/legal record. The forensic watch-items are quality-of-earnings (unbilled AR, OCF conversion) and governance (Mas-family related parties), not enforcement.

---

Phase D — Project & stress-test

Forward Projection (FY2026 / FY2027 / FY2028, adjusted diluted EPS)

Anchor = FY2025 actual adjusted diluted EPS $6.55, FY2026 management guidance $8.79 midpoint, Q1'26 run-rate $1.39 (+173%), backlog $20.3B (+28% YoY). Shares ~78.9M, roughly flat (modest buyback offsetting SBC/earn-out dilutionDilutionIssuing new shares, so each existing share owns a smaller slice of the same company.).

Conflict flagged: one data source shows a stale "analyst consensus $6.76" that sits below FY25 actual and far below the raised guidance — almost certainly a pre-Q1 / not-yet-updated print. I weight management's raised $8.79 guidance as the better-grounded FY26 anchor and treat $6.76 as stale. Surfacing both per provenance rules.

Reasoning lines (all, arithmetic shown):

• FY26 base $8.79 = management midpoint (revenue $17.5B × ~8.6% adj-EBITDA margin → ~$1.5B EBITDA → ~$8.79 adj EPS).
• FY27 base $10.6 = $8.79 × 1.21 (revenue +18%, ~20bps margin lift) — backlog already supports the top line.
• FY28 base $12.4 = $10.6 × 1.17.
• Bear path holds EPS roughly flat $8–8.5 — the scenario where the data-center wave does not fully offset the clean-energy-credit air-pocket and working capital keeps starving cash.

Forecast NOT logged — per --watchlist rules, the Brier our model create step is skipped in breadth mode (only logged when genuinely committing the base case; that is the human-gated our position log step). Suggested forecast for later: "MTZ FY26 (Dec 2026) adjusted diluted EPS ≥ $8.50, p≈0.70, resolves 2027-02-28."

Bull vs Bear

Bull case. MasTec is the purest non-utility, non-equipment-OEM way to own the AI-electricity buildout. The demand triad is structural and mutually reinforcing: (1) data-center load (peak US demand +26% by 2035, data-center demand 5x by 2035 per Deloitte ) needs (2) new generation — renewables were 93% of 2025 capacity adds — and (3) massive T&D expansion + (4) gas pipeline to firm it. MasTec self-performs all four. The proof is in the numbers, not the story: record $20.3B backlog (+28% YoY, CE&I +65%), Q1 revenue +34%, adjusted EPS +173%, guidance raised 34%. Margins are still below normalized (6.7% Q1 vs an achievable 9–10%), so there's operating leverage left. Founder-led, ~23% insider-owned, de-levered to ~1.7x, resuming buybacks. If the cycle runs 3–5 more years, EPS compounds 15–20% and the stock grows into its multiple.

Bear case (2–3 permanent-impairment risks).

1. It's a cyclical contractor priced as a secular compounder. At ~40x forward / 26x EV/EBITDA, MTZ prices durable 20%+ growth. The business has 7–15% segment margins, cancelable MSAs, and a history of losses (FY23). The first guidance cut or backlog stall re-rates it toward EMCOR's 26x — ~35% multiple compression even before any earnings miss.
2. Policy/renewables air-pocket. OBBBA (July 2025) accelerated the IRA clean-energy tax-credit phaseout. CE&I is the largest and fastest-growing segment, heavily renewables-linked. If the data-center demand thesis does not seamlessly replace credit-driven renewable demand, CE&I growth + backlog could disappoint sharply — and CE&I backlog (+65%) is what the bulls are paying for.
3. Quality of earnings / cash conversion. OCF fell to $546M while net income rose; FCF ~$285M is <1% of revenue and far below earnings; unbilled AR + DSO are creeping up. A working-capital reckoning (or POC-estimate revision) would expose the gap between adjusted EPS and cash.

Pre-mortem (it's Dec 2027, the thesis broke — what happened?): Hyperscaler capex digestion + an IRA-credit-driven renewables cliff hit CE&I in late 2026; a couple of large fixed-price data-center/pipeline jobs took cost overruns (POC true-ups), denting margins; backlog growth stalled; management cut FY27 guidance; the stock de-rated from 40x to ~22x on lower EPS — a 45–55% drawdown despite the business still being fine, because the expectations (not the company) broke.

Are multiples too high? Yes, on any normalized-margin / mid-cycle basis — but "too high" can persist for years while a super-cycle runs. This is an expectations-risk name, not an accounting-fraud name.

Contrarian view (what the market refuses to see): The bull consensus treats data-center demand and clean-energy demand as additive tailwinds. The 10-K's own language hints they may be partly substitutive — OBBBA cuts renewable credits while data-center electricity demand rises. If data-center load mostly backfills a renewables air-pocket rather than stacking on top, MasTec's growth is more like high-teens than the +30% the stock is extrapolating — still good, but not 40x-P/E good.

Devil's Advocate (short-seller)

Dismantling the bull case.

• Where revenue is concentrated / what breaks it: top-10 customers = 34%; the growth is increasingly data-center-tied (CE&I + Power Delivery). A hyperscaler capex pause — the single most-crowded long thesis in the market — would hit MasTec's hottest backlog. And MSAs are cancelable: the $20.3B backlog can shrink without a "cancellation event" headline.
• Why the moat is weaker than bulls think: it's a scale-and-labor moat, not a structural one. In a downturn, every E&C contractor (Quanta, EMCOR, Primoris, plus regional players) chases the same shrinking project pool, and margins — already only 7–15% — compress fast. There is no pricing power without scarcity.
• Most dangerous competitor bulls underestimate: Quanta Services (PWR) — larger, higher-multiple, arguably better-positioned in the highest-value T&D + utility work, and it can outbid on the marquee data-center power jobs. MasTec is the #2, and #2 in a commoditizing services market is where margin gets squeezed first.
• Worst capital-allocation / governance items: the Mas-family related-party web — $77.6M of FY25 revenue (3x in two years) building the family's Inter Miami stadium with a $37.5M related-party receivable, plus aircraft leases and CCI rentals. Serial M&A (4+ deals/yr) creates integration + goodwill-impairment risk and makes organic vs acquired growth hard to disentangle (though Q1'26 organic +29% is reassuring). Classified board + anti-takeover provisions entrench the family.
• Assumptions that must hold for $380: ~20%+ revenue growth for 2–3 more years, margin expansion to ~9–10%, no working-capital/POC blowup, no renewables policy cliff, no hyperscaler capex pause. That's a lot of things going right, priced as the base case.
• If growth disappoints by 20–30%: FY27 EPS lands ~$8 instead of ~$10.6, the multiple compresses to ~22–25x, and the stock is $180–220 — roughly 45–55% downside. The valuation gives almost no margin of safety.
• The single scenario that permanently impairs: a multi-year AI-capex digestion coinciding with the IRA-credit renewables cliff — the data-center and clean-energy legs sag together, backlog converts at lower margins, and a fixed-price project loss forces a POC restatement. Plausibility: low-to-moderate in the next 12 months (the backlog and Q1 print argue against an imminent break), moderate on a 2–3yr view given how much perfection is priced.

Management Questions (ordered by information value)

1. Of the record $20.3B backlog, what share is data-center-related, and how much of that is under cancelable MSAs vs firm fixed-scope contracts? (The whole bull thesis hinges on backlog durability.)
2. OBBBA accelerated the IRA clean-energy credit phaseout — quantify the revenue at risk in CE&I, and is incremental data-center/generation demand additive to renewables demand or substitutive for it?
3. FY25 operating cash flow fell to $546M while net income rose — when does OCF reconverge toward EBITDA, and what's the FY26 free-cash-flow target?
4. Unbilled receivables grew to $2.0B and DSO to 65 days — what is driving the working-capital build, and where do you expect DSO to normalize?
5. Q1 Pipeline EBITDA margin hit 21.2% — how much of that is sustainable run-rate vs project-mix timing, and what's the normalized through-cycle Pipeline margin?
6. Consolidated adjusted EBITDA margin is ~8% with segment margins 7–15% — what's the realistic normalized-margin ceiling, and what's the bridge to get there?
7. What is the hyperscaler/data-center customer concentration within Power Delivery and CE&I, and how exposed is the backlog to a single customer's capex deferral?
8. Capital allocation: with leverage ~1.7x and buybacks resumed, how do you prioritize M&A vs buybacks vs balance-sheet flexibility from here?
9. On the Inter Miami construction work ($77.6M FY25, $37.5M receivable) — what governance/independent-review process covers Mas-family related-party transactions, and what collection terms apply to that receivable?
10. You've done 4+ acquisitions a year — how do you measure integration success, and what's the goodwill-impairment sensitivity if organic growth slows?
11. How much of FY26's guided growth is organic vs acquired, and what's the organic backlog-to-revenue conversion assumption?
12. Skilled-labor availability is the binding industry constraint — how are wage inflation and crew availability trending, and how is that reflected in bid margins?
13. Transformer/high-voltage equipment lead times remain extended — is supply of long-lead gear a constraint on Power Delivery backlog conversion?
14. How does the competitive dynamic vs Quanta on large turnkey data-center power packages affect your win-rate and bid discipline?
15. What end-market would you most want to reduce exposure to over the next three years, and why?

---