A place in Britain's electricity connection queue has become an asset with option value. A data-centre developer can reserve access while it searches for customers, finance, equipment and planning consent. When the reservation is cheap relative to the value of future power, applying early is rational even if the project is uncertain. Repeated across developers and sites, that logic produces a queue much larger than the capacity likely to be built.
Ofgem now proposes to change the option's price. The regulator's data-centre commitment fee and project milestones are meant to make applicants put capital and evidence behind their reservation. The reform could give investors a more credible pipeline. It cannot answer the separate question of how quickly Britain can build the generation, substations and wires that viable projects still require.
A free queue position can multiply demand
Wired's report describes a queue crowded by projects whose power requests may never become operating facilities. Ofgem provides the scale: contracted demand offers rose from 41 GW to 125 GW between November 2024 and June 2025, and its consultation references roughly 73 GW of data-centre demand. The figures are applications and offers, not a forecast of energized server halls.
That distinction is central. A connection reservation preserves upside if AI demand, finance and land approvals align. The developer can abandon it if they do not. Much of the downside of holding a speculative place falls on the system: network planners may design around inflated demand, and ready projects can appear to sit behind capacity that will never be used.
The queue therefore aggregates two things that look identical on paper: committed capital projects and inexpensive claims on a possible future. Counting megawatts without estimating their probability of delivery overstates both electricity demand and investable data-centre supply.
The fee prices delay; milestones price evidence
Ofgem's formal consultation proposes two linked screens. A relevant project would secure a commitment fee from acceptance of its offer until it is ready to connect; the fee would not be returned if the project failed to progress under the rules. Separate milestones would require evidence of a credible end user, procurement of long-lead electrical equipment such as switchgear, and the financial and technical capacity to develop the site.
The fee changes the carrying cost of waiting. A developer keeping multiple uncertain sites must tie up more capital and risks losing it. The milestones change the information set. An end-user contract tests demand, an equipment order tests schedule and spend, and evidence of capability tests whether the applicant can execute beyond obtaining a queue number.
Neither screen is perfect. A well-capitalized group can afford to hold options that a smaller developer cannot, even when the smaller project is better located. A credible project can also lack a final customer precisely because the connection date is uncertain. If milestones are too early or rigid, the reform may select balance-sheet strength rather than deliverability.
Curation frees forecasts, not electrons
The National Energy System Operator says Britain's former connection queue exceeded 700 GW across technologies and was around four times the capacity needed for 2030. Its reform summary describes a reordered pipeline intended to prioritize ready projects and strategic needs. In August, NESO also argued that a queue that could never be delivered harmed certainty for investment.
Removing a phantom data centre can release a position in planning models and let a viable project advance. It does not instantly release physical power at the same location. Network constraints can be regional; a cancelled request in one zone may not help a project elsewhere. New substations, transmission reinforcement and generation retain their own lead times, permits and costs.
This is the analytical boundary investors should preserve. Queue reform improves allocation and information. Network investment expands the feasible set. Treating the first as a substitute for the second would turn a useful administrative change into an unsupported capacity forecast.
Conversion should replace queue size as the score
The regulator's counterargument is strong. Leaving speculative projects untouched also favors incumbents that accumulated early positions and imposes hidden costs on every project waiting behind them. A refundable-unless-idle fee plus evidence milestones could be more neutral than an opaque first-come system. Ofgem's demand-reform update explicitly separates curation, strategic planning and new connection arrangements, which acknowledges that screening is only one pillar.
Success should therefore be measured as conversion rather than deletion. Useful indicators include the share of queued megawatts reaching each milestone, time from offer to energization, fees forfeited, projects displaced despite later delivery, regional connection times and the difference between contracted and operating demand. Falling queue capacity alone could mean better screening or lost viable investment.
The thesis would strengthen if connection times fall and a larger share of the remaining pipeline secures customers, equipment and financing. It would weaken if wealthy applicants simply pay to warehouse positions or if cancellations rise without more energized capacity. Britain is pricing the right to wait. The return on that reform depends on whether a cleaner queue becomes a faster physical grid.