Grid delivery services are the contractors and specialty electrical builders that turn utility, data-center, industrial, power-generation, and interconnection demand into transmission lines, distribution upgrades, substations, generation tie-ins, electrical rooms, and site power connections after a buyer funds a defined scope. The business mechanism is backlog and RPO conversion into segment margin, DSO control, operating cash flow, and free cash flow after labor, equipment, materials, capex, and working-capital needs.
What the stack is: project-delivery companies that supply crews, supervisors, safety systems, trenching and lifting equipment, electrical installation, civil work, project controls, procurement, change-order management, and customer billing for power infrastructure.
What it does: the stack converts a funded scope into built assets: poles and wires, feeders, substations, switchyards, generation tie-ins, underground utility routes, data-center electrical rooms, and industrial power connections.
Main operating pieces: craft labor, union and non-union crews, bucket trucks, cranes, boring equipment, transformers and switchgear supplied by others, procurement teams, subcontractors, estimators, schedulers, safety programs, claims teams, and receivables collection.
Where it sits: on utility rights of way, at substations and generation interconnects, under streets, inside data-center and industrial campuses, and at customer electrical rooms after the utility, developer, generator, or public agency has authorized the work.
How the theme uses it: power scarcity creates work only after load requests, rate recovery, customer funding, generator awards, or industrial capex become a signed and staffed project. Contractors get paid as backlog or remaining performance obligations convert into revenue, margin, and cash.
Terms used later: backlog is contracted or expected work still to be performed; remaining performance obligations, or RPO, are enforceable contracted revenue still to be recognized; DSO means days sales outstanding and shows billing/collection speed; fixed-price work puts cost-overrun risk on the contractor; change orders and claims are requests to recover scope, cost, or schedule changes.
Report boundary: this node is a tactical report layer. It ranks the current value-chain basket, setup labels, confirmation triggers, invalidation levels, and chart provenance. Durable research, claim IDs, raw source registries, and sector thesis maintenance stay in the linked knowledge pages. Price and volume context comes from read-only discovery daily_ohlc through 2026-06-18.
Current Setup
Execution conversion gateFunded load matters only if backlog turns into margin and cash.
Contractors do not earn from load forecasts alone. The proof is signed scope, staffed work, stable gross margin, controlled DSO, clean change-order recovery, and operating cash flow after capex and working capital.
Positive proofPWR, MYRG, AGX, and MTZ now anchor the delivery basket.
PWR has scale, MYRG has smaller T&D purity, AGX has power EPC concentration, and MTZ needs cash-conversion proof.
Primary constraintCost, labor, valuation, and stale source coverage can absorb demand.
AGX must prove FY2026 margin durability, MTZ needs DSO/FCF proof, and PRIM needs a lane refresh after Q1.
Demand sourceUtility, data center, generation
Funded scopeAward, MSA, RPO, backlog
ExecutionLabor, equipment, change orders
Shareholder proofMargin, DSO, FCF
Power demand does not become contractor earnings until utilities, data-center developers, power generators, industrial customers, and public agencies convert planning into funded scopes. The node confirms only when the work can be permitted, staffed, built, billed, and collected without margin erosion or working-capital stress.
The strongest local evidence is company-specific. PWR has the clearest grid-delivery route, with Electric at 82.1% of Q1 revenue, more than $40 billion of Electric backlog, raised FY2026 adjusted EBITDA guidance, improved DSO, and a named NiSource large-load project that still needs economics and timing detail. MYRG has direct T&D and C&I electrical exposure, 13.4% Q1 gross margin, net cash, and $2.54 billion of backlog expected within 12 months. AGX adds a fresh power EPC lane: FY2026 revenue was $944.6 million, Power was 80.1% of revenue, Power backlog was more than $2.7 billion, consolidated backlog was $2.929 billion, gross margin was 20.5%, and cash/investments were $895.0 million with no debt. MTZ has large Power Delivery and CE&I backlog plus raised FY2026 guidance, but DSO and free cash flow need improvement. EME and FIX add data-center electrical/mechanical delivery routes, while IESC, STRL, PRIM, and DY remain lower-ranked until source depth or direct grid-relief evidence improves. The next positive proof is Q2 and H2 evidence that backlog converts with stable segment margins, clean billing, and operating cash flow.
The setup weakens if demand stays in planning studies, interconnection queues, LOIs, non-minimum MSAs, or cancellable backlog, or if funded work converts into low-margin fixed-price revenue. Labor shortages, subcontractor pricing, fuel, freight, metals, tariffs, permitting delays, claims, change orders, closeout disputes, contract assets, retainage, DSO, refinancing cost, and capex can absorb the economics before shareholders see cash. AGX has a higher valuation and cash-quality proof burden after a major rerating because FY2026 operating cash flow benefited from $214.7 million of contract-liability inflow and fixed-price EPC margins can normalize. Newly added EME, FIX, STRL, and DY need current source checks before stronger operating claims; PRIM remains the largest stale-source row and should be refreshed before relying on it as a current operating read.
Paper-led rerank uses the 1GT grid-upgrade paper and the SemiAnalysis BTM/grid-headroom paper as prior updates. Local discovery shows active basket tickers with daily_ohlc coverage through 2026-07-02; newly added tickers have right-rail chart routes but no refreshed static setup labels unless already present. Ranking uses source-backed bottleneck exposure and economic-capture evidence first, not recent price performance.
Basket
This basket is inherited from node-data.json and report.json; it is not re-ranked here. The order weights physical grid delivery first: bulk T&D, substations, interconnections, generation tie-ins, and customer-site electrical work. PWR, MYRG, AGX, and MTZ are the core sleeve; EME, FIX, IESC, and STRL are option rows tied more to data-center or mission-critical project delivery; PRIM and DY stay in watch because the source route is stale or less direct to bulk grid relief.
Core row and clearest grid-delivery route across transmission, distribution, substations, generation tie-ins, and large-load interconnections.
Market cap$100.3B
Next earningsJul 30 est.
Proof metric$48.47B backlog
Role in stack
PWR builds transmission, distribution, substations, generation tie-ins, large-load interconnections, utility hardening, and related infrastructure for utilities, developers, industrial customers, and large-load buyers. The economics come from funded work becoming Electric revenue, segment margin, DSO control, and free cash flow.
Revenue mix
Q1 2026 revenue was $7.874787B. Electric Infrastructure Solutions was $6.469B, or 82.1% of revenue, and Underground Utility and Infrastructure Solutions was $1.406B. Electric revenue grew 30.8%.
Proof burden
Total backlog was $48.47B, Electric backlog was just over $40B, and RPO was $26.24B. The proof burden is Electric backlog conversion, DSO near the improved 61-day level, free-cash-flow delivery against guidance, and clearer NiSource large-load project economics.
Core row with a smaller but cleaner electrical-contractor route to T&D, substations, utility infrastructure, and C&I electrical execution.
Market cap$6.7B
Next earningsJul 29 est.
Proof metric$2.84B backlog
Role in stack
MYRG sells transmission, distribution, substation, utility-infrastructure, and commercial or industrial electrical work. The node route is project backlog becoming revenue, gross margin, operating cash flow, and net-cash balance-sheet support.
Revenue mix
Q1 2026 revenue was about $1.00B. Transmission and Distribution was $541.0M, or 54.1% of revenue, and Commercial and Industrial was $459.4M, or 45.9%.
Proof burden
Gross margin was 13.4%, backlog was $2.84B, and $2.54B of backlog was expected within twelve months. The proof burden is Q2 margin repeatability, clean contract assets and retention, project closeouts, claims discipline, and labor availability.
Core power EPC and project-execution row tied to generation, industrial, and teledata projects rather than regulated grid ownership.
Market cap$9.9B
Next earningsLate summer
Proof metric$2.929B backlog
Role in stack
AGX provides power EPC, commissioning, maintenance, development, and consulting for power generation and related industrial or teledata projects. The economics come from project starts, RUPO and backlog conversion, Power margin, and cash collection.
Revenue mix
FY2026 revenue was $944.6M. Power generated 80.1% of revenue and held more than $2.7B of backlog; Industrial generated 17.7% of revenue, and Teledata generated 2.2%.
Proof burden
Consolidated backlog was $2.929B, gross margin was 20.5%, operating cash flow was $414.7M, and cash/investments were $895.0M with no debt. The proof burden is RUPO conversion, Power margin durability, fixed-price cost control, customer starts, and contract-liability normalization.
Core scale contractor with power delivery, clean-energy, civil, communications, and pipeline work tied to utility and data-center construction.
Market cap$29.5B
Next earningsJul 30 est.
Proof metric$20.328B backlog
Role in stack
MTZ builds power delivery, clean-energy, civil, energy, communications, and pipeline projects for utilities, energy customers, data-center-linked buyers, and infrastructure owners. The route is scale backlog moving into EBITDA and free cash flow without higher billing drag.
Revenue mix
Q1 2026 revenue was $3.829B, up 34.5%. Communications was $802.1M, Clean Energy and Infrastructure was $1.329B, Power Delivery was $1.046B, and Pipeline Infrastructure was $682.5M.
Proof burden
Adjusted EBITDA was $283.6M, backlog was $20.328B, Power Delivery backlog was $6.222B, and CE&I backlog was $7.279B. The proof burden is DSO at 72, $11.9M of free cash flow versus adjusted EBITDA, and about 44% of backlog under non-minimum MSA-like agreements.
Option row for electrical and mechanical project delivery at data-center, mission-critical, and network sites.
Market cap$34.4B
Next earningsNot confirmed
Proof metricRecord RPO
Role in stack
EME supplies U.S. electrical and mechanical construction and services for customer-site power, data-center, network, and mission-critical work. The conversion route is project RPO becoming revenue, U.S. Electrical and Mechanical margin, receivable collection, and service pull-through.
Revenue mix
The EME lane supports U.S. electrical and mechanical construction, services, record RPO, and data-center or network project exposure. The reviewed node package does not add a fresh segment bridge in this HTML.
Proof burden
The proof burden is U.S. Electrical and Mechanical margin, RPO conversion, receivable collection, and data-center or network awards that convert into cash instead of only higher activity.
Option row for mechanical, electrical, and technology-building delivery, especially data-center and technology-facility projects.
Market cap$61.3B
Next earningsNot confirmed
Proof metricBacklog + FCF
Role in stack
FIX delivers mechanical, electrical, technology, construction, and service work for buildings and customer sites. In this node it is a data-center and technology-building delivery route, with less direct transmission and substation exposure than PWR or MYRG.
Revenue mix
The FIX lane supports mechanical, electrical, and technology construction and service work with strong backlog and cash evidence. The node package does not provide a fresh segment revenue bridge.
Proof burden
The proof burden is technology and data-center revenue, backlog margin, free cash flow, and customer concentration detail that shows the building-delivery route is not just a temporary project spike.
Option customer-site and data-center electrical delivery row through Communications, Infrastructure Solutions, C&I, and related services.
Market cap$13.0B
Next earningsAug 7 est.
Proof metric$3.86B backlog
Role in stack
IESC supplies electrical, mechanical, technology-infrastructure, and industrial services that connect AI and power-intensive customer facilities. It is customer-site delivery, not utility-regulated grid ownership or rate-base recovery.
Revenue mix
Q2 FY2026 revenue was $974.3M. Communications was $367.7M, Residential was $287.6M, Infrastructure Solutions was $192.4M, and Commercial and Industrial was $126.5M.
Proof burden
Communications revenue grew 34.7%, Infrastructure Solutions grew 63.6%, backlog was $3.86B, and RPO was $2.35B. The proof burden is fixed-price exposure, percentage-of-completion accounting, backlog quality, customer concentration, residential margin drag, and Gulf Island integration.
Option row for campus, site, and mission-critical infrastructure that helps customer facilities become power-ready.
Market cap$21.5B
Next earningsNot confirmed
Proof metricSigned backlog
Role in stack
STRL fits sitework, E-Infrastructure, and mission-critical infrastructure that prepares data-center or industrial campuses for power and construction. The route is funded site work becoming signed backlog, margin, and cash conversion.
Revenue mix
The STRL lane supports E-Infrastructure, mission-critical site work, and construction services, with fixed-price and unsigned-award quality as important gates. The node package keeps it below the direct bulk-grid contractors.
Proof burden
The proof burden is signed backlog quality, E-Infrastructure margin, cash conversion, repeat mission-critical awards, and clear treatment of unsigned awards before the row carries more weight.
Watch row for utilities and energy-infrastructure construction with a stale local lane and a post-lane guidance change.
Market cap$4.8B
Next earningsNot confirmed
Proof metric$11.6B backlog
Role in stack
PRIM provides gas, electric, communications, power, renewables, storage, and industrial infrastructure services. It remains in the basket because the business touches grid and energy delivery, but the current operating read needs a lane refresh.
Revenue mix
Local lane baseline is FY2025: revenue was $7.57B, Energy was $5.02B, and Utilities was $2.69B before eliminations. A bounded official Q1 check shows total backlog at $11.6B, Utilities backlog at $6.9B, and Energy backlog at $4.7B.
Proof burden
FY2025 operating cash flow was $470.4M versus $129.9M of capex. The May 5 Q1 release lowered 2026 adjusted EBITDA guidance to $480M-$500M and included PayneCrest, so Q1 revenue, Energy margin, renewables cost, integration, cash conversion, and guidance need refresh.
Watch row where acquired building-systems exposure adds some critical-facility route, but telecom and fiber remain the main setup.
Market cap$13.1B
Next earningsNot confirmed
Proof metricFCF deleveraging
Role in stack
DY supplies telecom, utility, fiber, and acquired building-systems construction services. In this node it is a watch row for critical-facility and utility-adjacent work, with customer capex and acquisition integration controlling the conversion path.
Revenue mix
The DY lane supports telecom, utility, fiber, and acquired building-systems exposure, with leverage and cash conversion central after acquisitions. The node package does not provide a fresh segment revenue bridge.
Proof burden
The proof burden is Building Systems margins, retention, free-cash-flow deleveraging, customer capex durability, and evidence that the building-systems route can matter alongside the larger telecom and fiber base.
Market caps use the read-only discovery method referenced in report.json: latest 2026-07-02 close multiplied by stored weighted shares outstanding for PWR, MYRG, AGX, MTZ, EME, FIX, IESC, STRL, PRIM, and DY. Earnings dates use the existing node source trail where checked; unverified rows stay marked Not confirmed. Card wording is routed to the linked security lanes and the node's ordered basket_rows.
What Confirms Or Weakens
AreaWhat confirmsWhat weakens or invalidatesWatch next
01Node thesis
Demand becomes funded work
What confirms
Utility, data-center, industrial, generator, and public-sector buyers turn load plans into funded awards, RPO, backlog, or MSAs with visible minimums.
What weakens or invalidates
Demand remains in planning studies, interconnection queues, LOIs, cancellable backlog, or non-minimum MSAs without funded work packages.
Watch next
Award language
Customer funding
RPO and backlog quality
02Economics mechanism
Backlog converts cleanly
What confirms
PWR Electric, AGX Power EPC, MYRG T&D/C&I, MTZ Power Delivery/CE&I, IESC Communications/Infrastructure Solutions, and PRIM Utilities convert backlog or RPO into revenue with stable or improving margins.
What weakens or invalidates
Revenue growth comes with fixed-price losses, Power margin reversion, labor or subcontractor overruns, tariff pressure, project delays, claims, closeout disputes, or backlog cancellation.
Watch next
Segment margin bridges
Book-to-bill
RUPO conversion
Project closeouts
03Customer and funding
Buyer can pay
What confirms
Utilities, hyperscalers, industrial customers, generators, and public agencies disclose funded scopes, rate recovery, customer contributions, or capex budgets that match contractor awards.
What weakens or invalidates
Customers delay financial close, cancel load, defer capex, lose rate-recovery support, or push contract risk back to contractors without price protection.
Adjusted EBITDA grows faster than cash, DSO rises, contract assets build, contract-liability cash reverses, free cash flow lags, debt or capex absorbs growth, or refinancing costs rise.
Watch next
PWR DSO and FCF guide
AGX contract liabilities
MYRG operating cash flow
MTZ DSO and FCF
05Operating constraint
Execution capacity holds
What confirms
Skilled labor, subcontractor availability, equipment fleets, procurement, permitting, and safety execution support revenue growth without claims or margin leakage.
Local discovery rows remain current through 2026-06-18, the AGX lane covers the 2026-06-21 paired research/profile output, and no new filing, earnings release, contract, guidance change, financing update, or rate-order evidence has arrived after the cited sources.
What weakens or invalidates
A new trading session, FY2027 quarterly release, 10-Q, investor-day deck, contract award, guidance change, debt update, AGX margin/cash update, or PRIM lane refresh changes the setup before the node is refreshed.
AGX latest profile output: profile-led read for power EPC exposure, backlog, margin, cash, no debt, risks, and next evidence.
AGX latest research output: research claims and evidence for Power revenue share, backlog growth, gross margin, operating cash flow, contract-liability timing, fixed-price risk, valuation, and discovery-backed market data.
PWR Q1 2026 results release: Official IR release used to cross-check Electric revenue, backlog, guidance, DSO, and large-load project detail.
AGX official facts are routed through the 2026-06-21 filed AGX lane and raw profile/research outputs, which cite the FY2026 earnings 8-K, FY2026 10-K, April 2026 capital allocation 8-K, proxy, and deterministic local discovery rows.
MYRG Q1 2026 results exhibit: SEC-filed exhibit used to cross-check revenue mix, backlog, cash flow, and balance-sheet facts.
MTZ Q1 2026 results release: Official IR release used to cross-check revenue, segment backlog, adjusted EBITDA, guidance, DSO, and free-cash-flow gates. The local MTZ lane contains a conflicting Q1 revenue figure, so the page uses the official release.
IESC fiscal Q2 2026 results release: Official IR release used to cross-check customer-site electrical exposure, segment mix, backlog, and RPO.
PRIM Q1 2026 results release: Official IR release used only as a bounded stale-lane check for updated backlog mix, lowered 2026 adjusted EBITDA guidance, and PayneCrest.
AGX next earnings is listed as a late-summer FY2027 quarterly catalyst from the AGX research output, based on its fiscal cadence after the July 31, 2026 quarter close; no separate external date was added.
Barchart PRIM earnings estimates, checked 2026-06-13: no usable next date shown. Nasdaq and Earnings Whispers checks did not produce a confirmed date, so PRIM is marked Not confirmed.
Discovery And Chart Provenance
duckdb -readonly ../discovery/data/discovery.duckdb showed daily_ohlc rows through 2026-06-18 for PWR, AGX, MYRG, MTZ, IESC, and PRIM; AGX coverage runs from 2021-05-11 to 2026-06-18.
Read-only DuckDB market-cap check used each ticker's 2026-06-18 close multiplied by stored weighted shares outstanding in the local discovery store. AGX market cap was about $10.36B using 14,019,247 weighted shares.
The right-rail chart route is /api/securities/{ticker}/chart?frequency=weekly&window=3y&as_of=latest. It reads daily_ohlc and returns weekly bars using first open, maximum high, minimum low, final close, and summed volume.
The API chart package is expected to expose 20w_ema, 100w_ema when history is long enough, weekly_volume, and 20w_average_volume. Moving averages are computed from weekly closes using full available weekly history before visible-window clipping.
Visible chart horizon is three years through 2026-06-18. Static setup thresholds from the 2026-05-22 node-data package are stale versus the current local price rows and must be refreshed before use as current trading evidence.
Review validation previously started the local report API, but a sibling curl to the PWR chart route could not connect to 127.0.0.1:8765 in this sandbox. This update records the route and service assumptions but does not embed OHLC arrays.
discovery lineage-status --dataset daily_ohlc can lag the latest stored OHLC rows, so use the page's chart data as local price and volume context until ingestion lineage is reconciled.
Known Gaps
Discovery supports price, volume, weekly OHLC, EMA, and coverage facts only. It does not support causal price-action claims without a filing, earnings release, contract, rate recovery, financing, or guidance update.
Local discovery status previously showed 13F holdings rows were empty for the original basket. This update does not add holder-concentration claims for AGX or peers.
PWR has the freshest scaled grid-delivery evidence, but NiSource large-load economics, timing, risk-sharing, contract terms, debt maturity, acquisition integration, claims, and change-order risk still need follow-up.
AGX is power EPC/project execution capacity. The live proof burden is backlog/RUPO conversion, Power margin durability, fixed-price cost control, customer starts, contract variations, working-capital normalization, and valuation after a major rerating.
MYRG needs follow-up on Valley and Comet acquisition economics, integration risk, and whether backlog growth converts into cash without higher working-capital drag.
MTZ has large Power Delivery and CE&I exposure, but local knowledge and the official release disagree on Q1 revenue. The official release is used here, and the MTZ lane should be reconciled.
IESC is adjacent customer-site and data-center electrical delivery; utility rate-base recovery and transmission ownership are outside this row. Exact data-center concentration is not disclosed.
PRIM local knowledge predates Q1 2026. The May 5 official release lowered 2026 adjusted EBITDA guidance and changed backlog mix, so the security lane should be refreshed before PRIM is treated as a current operating read.
The family generator scripts/build_power_scarcity_nodes.py still emits legacy basket and setup structures for this family. This page was updated as a static artifact only; do not rerun the stale generator over it without modernizing the generator first.