{"schema_version":"2026-09-05.topic-graph-v1","canonical_url":"https://www.manufacturing.ai/topics/data-centre-construction","topic":{"slug":"data-centre-construction","name":"Building the Shed","description":"A sourced reference collection on the buildings the accelerators go into — a North American market at 1.4% vacancy with four-fifths of its construction pipeline already pre-sold, what a watt of capacity costs to build across markets, and why the binding constraints have become electricity, consent and skilled labour rather than capital.","coverage_notes":"Nineteenth packet, and the one where source independence is weakest for a structural reason worth naming: there is no government statistical series for data-centre supply comparable to the mineral surveys used elsewhere in this index. Ten sources were consulted: five were retrieved and read, and five were surfaced and set aside with a stated reason. One read was opened in a browser after the automated fetcher proved unsuitable for that outlet; no paywall or challenge was circumvented, and a registration step guarding the current cost index was deliberately NOT completed, and that report is recorded as unread. Every quantitative claim on this page comes from one of two interested parties, and both interests are stated in the text rather than a footnote: the supply, vacancy, absorption and preleasing figures come from a commercial real-estate brokerage that earns fees on the transactions it is measuring, and the cost-per-watt benchmarks from a consultancy that sells cost management to the organisations building these facilities. They are used because the alternative is asserting nothing about the physical build-out at all. Two dating decisions matter. The cost benchmarks are from the 2025 edition of the index, because the 2025-2026 edition's figures sit inside a gated report that was not retrieved — a year of movement in the most inflationary part of construction is therefore missing, and the figures are labelled as a year old rather than presented as current. And the labour figures come from a different study by the same consultancy, reported by a trade outlet, not from the cost index, so they are attributed separately. The claim about local opposition is deliberately weakened to what the evidence supports: an interested and well-placed observer reports consent as a first-order siting constraint, which is not the same as a count of planning refusals establishing it. Named gaps, in order of value: the current edition of the cost index, gated; competing brokerages' supply and absorption series, whose absence means there is no way to tell from this page whether these figures reflect the market or one firm's view of it; utility and regional transmission interconnection queue data, which is the independent non-commercial measure against which 'contracted power' claims could be checked; county zoning and permitting records, which would replace a characterisation of local opposition with a count; and hyperscaler capital expenditure filings, which would be the demand-side check on a supply-side story. Not yet editor-reviewed; every assertion reads as reported.","primer":"Everything else in this index has to be installed somewhere. The somewhere has a vacancy rate, and in the first half of 2026 that rate was 1.4% across North American primary markets — down from 1.6% a year earlier, despite total supply growing 33.7% to 10,903 MW, because absorption kept pace with everything built.\n\nUnder-construction capacity reached 7,481.1 MW, up a quarter in six months and past the previous peak of 6,350.1 MW set in late 2024. Of that pipeline, 80.4% was already committed before completion, against 74.3% a year earlier. Less than 1,500 MW remained available across all primary markets — roughly six months of demand at the prevailing absorption rate. Building at a record rate, vacancy falling anyway, four-fifths of the pipeline pre-sold. That is not a market clearing; it is a queue.\n\nThe market-level figure understates the scarcity where demand actually is. Northern Virginia held 4,496.5 MW of inventory — the largest concentration in the world — against 10.8 MW available, while absorbing 467.7 MW in the half, about a third of all net absorption across primary markets. Hillsboro, Oregon: 491.4 MW of inventory, 1.1 MW available. Phoenix: 1,069.2 against 12.1. Competition has moved upstream, to securing power and land rather than leasing buildings.\n\nBuildings here are priced in dollars per watt rather than per square foot, and the unit is the point: the constraint is electrical, not spatial. A 2025 benchmark index spans roughly 60%, from $9.5 per watt in Charlotte, North Carolina through $9.8 in Phoenix and Columbus to $12.9 in New Jersey, $13.3 in Silicon Valley and $15.2 in Tokyo. The year's increase for a traditional air-cooled facility was put at 5.5%, itself down from 9.0% the year before.\n\nTurner & Townsend's 2025 analysis reports average construction costs 7 to 10 per cent higher for liquid-cooled US facilities than equivalent air-cooled ones of similar IT capacity. It is a construction-cost comparison, not a breakdown of total AI infrastructure investment or operating costs; the quoted premium alone does not establish where those wider costs sit.\n\nThe constraint capital cannot relieve is people. A global construction study reported in July 2026 found data centres the most constrained sector globally for contractor capacity, with over 70% of 112 markets analysed tightening or overstretched — against 79% of traditional sectors balanced or with spare capacity. Some 71% of markets reported labour shortages outright. The same study found input costs had stabilised, which locates the pressure precisely: on people, not materials. And the work is being pulled out of housing and commercial building, because technology work pays better.\n\nThe last constraint is consent. The market research states that local opposition has become a serious obstacle, with community resistance and zoning delays stalling projects, and that securing local approval is now as critical to siting as power and fibre. That would be a real change in an industry that spent two decades choosing sites on latency, power price and tax treatment — but it is a brokerage's characterisation of a planning environment, not a count of refusals, and the county records that would settle it were not retrieved.\n\nA word on whose numbers these are, because it matters more here than anywhere else in this index. There is no government statistical series for data-centre supply the way there is for minerals. The best available figures come from a commercial brokerage that earns fees on the transactions it measures, and the best cost benchmarks from a consultancy that sells cost management to the people building. Both are used because the alternative is asserting nothing. Neither is disinterested, and no competing brokerage's series was retrieved to check either.","primer_note":"Manufacturing.ai's own editorial synthesis, not evidence. Every factual statement in it is separately asserted and sourced in this response's concepts and events.","reviewed_through_date":null,"source_count":4,"inspected_source_count":4,"consulted_reference_count":6,"blocked_reference_count":2,"set_aside_reference_count":4},"figures":[{"kind":"time-series","title":"What a watt costs to build","caption":"Indicative construction cost per watt across selected markets. The spread between cheapest and dearest is roughly 60%.","sourceNote":"Turner & Townsend Data Centre Construction Cost Index, 2025 edition: survey-derived benchmarks in US dollars per watt, not measured averages of completed projects, published by a consultancy that sells cost management to firms building these facilities. A selection of the markets listed is plotted. The newer 2025-2026 edition's benchmarks sit behind a registration step that was not completed, so these figures are a year old.","unit":"Construction cost, US dollars per watt (2025 index)","points":[{"label":"Charlotte, NC","value":9.5,"display":"$9.5"},{"label":"Phoenix","value":9.8,"display":"$9.8"},{"label":"Atlanta","value":9.9,"display":"$9.9"},{"label":"Dublin","value":10,"display":"$10.0"},{"label":"Amsterdam","value":10.8,"display":"$10.8"},{"label":"New Jersey","value":12.9,"display":"$12.9"},{"label":"Silicon Valley","value":13.3,"display":"$13.3"},{"label":"Tokyo","value":15.2,"display":"$15.2"}]},{"kind":"time-series","title":"Supply, pipeline, and what is actually left","caption":"North American primary markets, first half of 2026. The available bar is smaller than a single half-year of absorption.","sourceNote":"CBRE North America Data Center Trends H1 2026: total primary-market supply of 10,903 MW, under-construction capacity of 7,481.1 MW, net absorption of 1,456.2 MW in the half, and less than 1,500 MW of available future capacity — plotted at that 1,500 MW upper bound, so the real figure is lower. Published by a commercial real-estate firm that earns fees on the transactions these figures describe.","unit":"North American primary markets, megawatts (H1 2026)","points":[{"label":"Available capacity","value":1500,"display":"<1,500 MW"},{"label":"Net absorption, the half","value":1456.2,"display":"1,456 MW"},{"label":"Under construction","value":7481.1,"display":"7,481 MW"},{"label":"Total supply","value":10903,"display":"10,903 MW"}]}],"blocked_references":[{"title":"AI is set to drive surging electricity demand from data centres while offering the potential to transform how the energy sector works","publisher":"International Energy Agency","url":"https://www.iea.org/news/ai-is-set-to-drive-surging-electricity-demand-from-data-centres-while-offering-the-potential-to-transform-how-the-energy-sector-works","source_type":"press_release","retrieval_status":"blocked_403","content_inspected":false,"published_at":null},{"title":"Energy and AI — Executive Summary","publisher":"International Energy Agency","url":"https://www.iea.org/reports/energy-and-ai/executive-summary","source_type":"other","retrieval_status":"blocked_403","content_inspected":false,"published_at":null}],"set_aside_references":[{"title":"Competing data-centre market reports from other brokerages","publisher":"JLL, Cushman & Wakefield and others","url":"https://www.jll.com/en-us/insights","source_type":"company_website","retrieval_status":"search_result_only","content_inspected":false},{"title":"County zoning and permitting records in major data-centre markets","publisher":"Local government planning authorities","url":"https://www.loudoun.gov/","source_type":"other","retrieval_status":"search_result_only","content_inspected":false},{"title":"Hyperscaler capital expenditure disclosures","publisher":"U.S. Securities and Exchange Commission (EDGAR)","url":"https://www.sec.gov/","source_type":"regulatory_filing","retrieval_status":"search_result_only","content_inspected":false},{"title":"Utility and regional transmission organisation interconnection queue data","publisher":"Federal Energy Regulatory Commission and regional transmission organisations","url":"https://www.ferc.gov/","source_type":"other","retrieval_status":"search_result_only","content_inspected":false}],"concepts":[{"slug":"ai-rack-power-density","name":"AI Rack Power Density","category":"component","canonical_url":"https://www.manufacturing.ai/topics/data-centre-construction#ai-rack-power-density","assertions":[]},{"slug":"cost-per-watt","name":"Cost Per Watt","category":"other","canonical_url":"https://www.manufacturing.ai/topics/data-centre-construction#cost-per-watt","assertions":[{"id":"63446650-3add-49d8-8588-9ebe0afa49f1","predicate":"concept.description","statement":"Turner & Townsend's 2025 analysis reports average construction costs 7 to 10 per cent higher for liquid-cooled data centres in the US than equivalent air-cooled facilities of similar IT capacity. This compares construction costs; it does not establish the distribution of total AI infrastructure or operating costs.","assessment":"reported","editor_reviewed":false,"reference_only_source_count":0,"evidence":[{"id":"462bc01e-c385-4ca5-a188-7d8a928d3d68","stance":"supports","is_primary_for_assertion":true,"origin_independence":"unknown","note":null,"quote":null,"source":{"title":"Data Centre Construction Cost Index 2025 — data centre cost trends","publisher":"Turner & Townsend","url":"https://reports.turnerandtownsend.com/data-centre-construction-cost-index-2025/data-centre-cost-trends","source_type":"company_website","retrieval_status":"fetched","content_inspected":true,"published_at":null,"documents":[]}}]},{"id":"86f49e29-0963-4c89-9291-394bf243c2a8","predicate":"concept.description","statement":"Data centres are benchmarked in dollars per watt because the constraint is electrical: a 2025 index spans $9.5 per watt in Charlotte to $15.2 in Tokyo, with the year's increase for a traditional air-cooled facility put at 5.5%, down from 9.0% the prior year.","assessment":"reported","editor_reviewed":false,"reference_only_source_count":0,"evidence":[{"id":"6ea634fe-51c7-4c1c-b8e7-071ad387fc5e","stance":"supports","is_primary_for_assertion":true,"origin_independence":"unknown","note":null,"quote":null,"source":{"title":"Data Centre Construction Cost Index 2025 — data centre cost trends","publisher":"Turner & Townsend","url":"https://reports.turnerandtownsend.com/data-centre-construction-cost-index-2025/data-centre-cost-trends","source_type":"company_website","retrieval_status":"fetched","content_inspected":true,"published_at":null,"documents":[]}}]}]},{"slug":"data-centre-electricity-demand","name":"Data Centre Electricity Demand","category":"other","canonical_url":"https://www.manufacturing.ai/topics/data-centre-construction#data-centre-electricity-demand","assertions":[]},{"slug":"grid-interconnection-constraint","name":"Grid Interconnection Constraint","category":"other","canonical_url":"https://www.manufacturing.ai/topics/data-centre-construction#grid-interconnection-constraint","assertions":[]},{"slug":"construction-labour-constraint","name":"The Construction Labour Constraint","category":"other","canonical_url":"https://www.manufacturing.ai/topics/data-centre-construction#construction-labour-constraint","assertions":[{"id":"42f50e0f-58ba-4ab3-85bc-5e5009f30d29","predicate":"concept.description","statement":"Market research states that local opposition has become a serious obstacle, with community resistance and zoning delays stalling projects, and that securing local approval is now as critical to site selection as power and fibre — an interested observer's characterisation, not a count of planning outcomes.","assessment":"reported","editor_reviewed":false,"reference_only_source_count":0,"evidence":[{"id":"ce29bffe-b4aa-461b-b927-4d47f8a25bec","stance":"supports","is_primary_for_assertion":true,"origin_independence":"unknown","note":null,"quote":null,"source":{"title":"North America Data Center Trends H1 2026","publisher":"CBRE","url":"https://www.cbre.com/insights/books/north-america-data-center-trends-h1-2026","source_type":"company_website","retrieval_status":"fetched","content_inspected":true,"published_at":null,"documents":[]}}]},{"id":"c47ab546-2ebc-4bdf-aaf1-fd81a71a214e","predicate":"concept.description","statement":"Data centres are the most constrained sector globally for contractor capacity, with over 70% of 112 markets analysed seeing tightening or overstretched capacity against 79% of traditional sectors balanced or with spare, and 71% of markets reporting labour shortages — while input costs had stabilised, locating the constraint in people rather than materials.","assessment":"reported","editor_reviewed":false,"reference_only_source_count":0,"evidence":[{"id":"a88cdcc6-2291-474e-88e3-f4265c5d33db","stance":"supports","is_primary_for_assertion":false,"origin_independence":"unknown","note":null,"quote":null,"source":{"title":"AI demand is increasing labor shortages and skills pressure in data center construction","publisher":"Data Center Dynamics","url":"https://www.datacenterdynamics.com/en/news/ai-demand-is-increasing-labor-shortages-and-skills-pressure-in-data-center-construction-report/","source_type":"journalism","retrieval_status":"fetched","content_inspected":true,"published_at":"2026-07-08","documents":[]}}]}]},{"slug":"data-centre-vacancy-and-preleasing","name":"Vacancy and Preleasing","category":"other","canonical_url":"https://www.manufacturing.ai/topics/data-centre-construction#data-centre-vacancy-and-preleasing","assertions":[{"id":"425a4358-0e25-4bd1-b72f-e4e2ee81901b","predicate":"concept.description","statement":"North American primary-market data-centre supply reached 10,903 MW in the first half of 2026 at a 1.4% vacancy rate, with 7,481.1 MW under construction of which 80.4% was already preleased and under 1,500 MW available — characterised by the publisher as about six months of demand.","assessment":"reported","editor_reviewed":false,"reference_only_source_count":0,"evidence":[{"id":"59536293-6adc-4ec3-8f1f-bb9b8fab0227","stance":"supports","is_primary_for_assertion":true,"origin_independence":"unknown","note":null,"quote":null,"source":{"title":"North America Data Center Trends H1 2026","publisher":"CBRE","url":"https://www.cbre.com/insights/books/north-america-data-center-trends-h1-2026","source_type":"company_website","retrieval_status":"fetched","content_inspected":true,"published_at":null,"documents":[]}}]},{"id":"fa7db51d-8ecb-4eab-a3bd-5512a1dc8c70","predicate":"concept.description","statement":"Availability is concentrated to near zero in the largest markets: Northern Virginia held 4,496.5 MW of inventory against 10.8 MW available while leading absorption at 467.7 MW, and Hillsboro held 491.4 MW against 1.1 MW available.","assessment":"reported","editor_reviewed":false,"reference_only_source_count":0,"evidence":[{"id":"b366c484-013c-4e94-88e0-c18cb44e6b64","stance":"supports","is_primary_for_assertion":true,"origin_independence":"unknown","note":null,"quote":null,"source":{"title":"North American Data Center Demand Continues to Outpace Supply Despite Record Construction Activity","publisher":"CBRE","url":"https://www.cbre.com/press-releases/north-american-data-center-demand-continues-to-outpace-supply-despite-record-construction-activity","source_type":"press_release","retrieval_status":"fetched","content_inspected":true,"published_at":"2026-08-27","documents":[]}}]}]}],"events":[{"slug":"record-data-centre-construction-h1-2026","title":"Record Construction, Record-Low Vacancy","category":"capacity_announcement","event_date":"2026-08-27","date_precision":"day","country":"United States","region":null,"canonical_url":"https://www.manufacturing.ai/topics/data-centre-construction#record-data-centre-construction-h1-2026","assertions":[{"id":"4c2fb57a-e4fd-44de-b814-734b0619c48d","predicate":"event.description","statement":"CBRE reported on 27 August 2026 that North American primary-market data-centre supply reached 10,903 MW in the first half of 2026 at a record-low 1.4% vacancy, with a record 7,481.1 MW under construction of which 80.4% was preleased, and under 1,500 MW available — about six months of demand.","assessment":"reported","editor_reviewed":false,"reference_only_source_count":0,"evidence":[{"id":"8915a0d1-06dc-4b69-aea4-25238dad1eaf","stance":"supports","is_primary_for_assertion":true,"origin_independence":"unknown","note":null,"quote":null,"source":{"title":"North American Data Center Demand Continues to Outpace Supply Despite Record Construction Activity","publisher":"CBRE","url":"https://www.cbre.com/press-releases/north-american-data-center-demand-continues-to-outpace-supply-despite-record-construction-activity","source_type":"press_release","retrieval_status":"fetched","content_inspected":true,"published_at":"2026-08-27","documents":[]}},{"id":"f63c1fd7-8ee3-4f31-8e58-8908b2526a23","stance":"supports","is_primary_for_assertion":true,"origin_independence":"unknown","note":null,"quote":null,"source":{"title":"North America Data Center Trends H1 2026","publisher":"CBRE","url":"https://www.cbre.com/insights/books/north-america-data-center-trends-h1-2026","source_type":"company_website","retrieval_status":"fetched","content_inspected":true,"published_at":null,"documents":[]}}]}]}],"policy":{"publication_model":"transparency","evidence_entries_may_share_a_source":true,"linked_documents_do_not_add_verification":true,"correction_intake_available":false,"assessment_meanings":{"primary-source confirmed":"Editor-reviewed against a first-party source.","cross-source corroborated":"Editor-confirmed across two independent origins.","reported":"Supported by the attached sources; not yet editor-reviewed.","disputed":"Sources conflict; competing accounts are both attached.","superseded":"Replaced by a newer assertion.","unverified":"No supporting evidence attached."},"evidence_eligibility":"`evidence[]` contains only sources whose content was actually retrieved and read (`content_inspected: true`). Sources that could not be retrieved are counted per-assertion in `reference_only_source_count`, listed for the whole collection in `consulted_references`, and never presented as evidence. `source_count` counts only sources an assertion actually rests on — it deliberately excludes `consulted_references`, so it is never inflated by material nobody could read.","topic_scope":"Assertions and source counts are limited to the topic's current packet lineage. Reusing a concept in another topic does not silently import that other topic's assertions.","rights":"Statements are Manufacturing.ai's own prose. Source excerpts are not redistributed; `quote` is present only where explicitly cleared.","attribution":"Cite as Manufacturing.ai, https://www.manufacturing.ai/topics/data-centre-construction","correction_url":"https://www.manufacturing.ai/methodology#corrections"}}