Summary

  • BDUK’s rolling review turns provider submissions into address-level evidence for Project Gigabit interventions. It does not turn every reported plan into a live connection.
  • The four labels—White, Grey, Black and Under Review—separate different combinations of present coverage, planned deployment, verification and competition. Only White premises are eligible for subsidy at this stage; eligibility still precedes procurement and delivery.
  • The September 2026 review closed on 1 October while BDUK was beginning its assessment. Until its postcode and UPRN outputs are published, the public cannot treat those submissions as a completed map or a record of completed builds.

The address that exists on paper

A property can occupy an awkward place in a broadband plan. A supplier may already serve it; may intend to extend a network there within three years; may have submitted a plan that still carries delivery risks; or may not yet have supplied enough evidence for the public authority to classify it. Those states can affect whether taxpayers fund another build. They do not tell a resident, by themselves, whether a connection can be ordered today.

That distinction is central to Building Digital UK’s National Rolling Open Market Review. The September 2026 review asked suppliers for existing and planned broadband coverage over a three-year horizon. BDUK says it runs the review in January, May and September, and uses submissions to inform Project Gigabit, GigaHubs, Gigabit Voucher eligibility and other interventions. The mechanism is a repeated market map for public procurement, not a nationwide service-availability test performed at every door.

The unit is the Unique Property Reference Number, or UPRN. In principle, an address-level record is more useful than a broad claim that a village or postcode is “covered”: it can show which premises a provider says its network reaches. But the precision of an address key should not be mistaken for certainty in the underlying claim. The network may be planned, not built; a build may need further permissions; or the evidence may still be incomplete.

Four labels, four different questions

The review uses four classifications. White means no qualifying gigabit infrastructure is available and none is expected within three years. Grey means one supplier has a qualifying network available or planned. Black means two or more suppliers have one available or planned. Under Review covers claims that need more checking: existing gigabit availability may not yet be verified, or a planned build may carry risks or evidence gaps.

Only premises classified as White are eligible for subsidy at this stage. That rule protects commercial investment from being displaced by public money where service already exists or a credible build is expected. Yet “not eligible now” does not mean “served now.” Under Review addresses remain subject to monitoring. BDUK may seek evidence of significant progress within the three-year period; if a commercial plan falls away, an address may later be mapped as eligible for intervention.

This is a useful distinction in a subsidy programme, but it creates a waiting state. A White address can move toward an intervention area, but an eligibility classification is not a signed contract. A contract is not a completed network. A network passing a property is not proof that a resident has ordered service, can obtain it at a reasonable price, or receives the promised performance. The map answers a narrower question: where might public support be justified after other evidence and review?

BDUK’s process makes that sequence explicit. Suppliers submit coverage data with supporting evidence; evaluators assess it and may ask questions; BDUK determines premise eligibility; suppliers receive feedback; and the review closes. A further Public Review and programme overlays can change which White premises enter a final intervention area. That final area can be smaller than the full set of addresses initially classed White.

A plan needs a delivery clock

The three-year horizon is an operational compromise. It gives commercial providers time to plan construction while allowing the state to identify areas likely to remain outside viable private deployment. It also gives an address time to sit in a status that is neither verified service nor immediate subsidy.

Ofcom’s separate 2025 planned-network report shows why the word “planned” needs a maturity signal. Based on provider plans as of January 2025, Ofcom estimated that full-fibre availability could reach 95% of residential premises by January 2028 if all plans were realised. When it counted only high-confidence plans—those with low-level design complete and funding committed—the estimate was 86%. The gap was not evidence that operators had failed; it showed how a projection changes when plans are filtered for maturity. It was also a separate dataset, a separate date and a different assessment from BDUK’s September 2026 review.

The same separation matters for current coverage. Ofcom’s Connected Nations 2025 report, using provider data from July 2025, put gigabit-capable availability at 87% of UK residential premises and full fibre at 78%. Those national figures help describe the market, but do not decide the status of one UPRN in a later BDUK review. Nor do they show take-up, affordability or whether an individual household’s connection works.

The September map is not published yet

The September 2026 review closed on 1 October. Its closure notice said BDUK had received a range of supplier submissions and would begin evaluating them; formal responses would follow. The notice said that postcode- and UPRN-level outputs would be published after assessment. It also said submissions would be formally assessed for England and Wales. As of that notice, the September file was a review in progress, not a public inventory of verified service or a final intervention map.

This timing matters because the May 2026 OMR outputs were published on 17 September, after the earlier review and assessment. The recurring cycle updates the evidence base, but an interval remains between a supplier’s statement, BDUK’s decision, the public release and a build on the ground. Without a published September output, it would be wrong to infer which areas gained eligibility, which providers’ plans were accepted, or how many premises were classified in each category.

The map is also a record with boundaries. BDUK’s published UPRN files carry a version and date, and the September request noted that an updated AddressBase epoch was not yet matched by the corresponding ONS release; BDUK warned that this could delay provision of baseline files. That is a data-supply caveat, not proof that the review failed. A useful public dataset should make its reference date and source version visible so that councils, providers and residents know which map they are reading.

The strongest version of this system would preserve the evidence state as well as the colour. A premise-level output should distinguish “built and verified” from “planned,” identify the review date, retain a confidence or evidence-stage field for plans, and show whether delivery depends on public support. It should also show when a plan was downgraded, withdrawn or superseded. Commercially sensitive engineering detail need not be public for the decision trail to be intelligible.

Sources