Ofgem’s "Curate" Consultation: What it means for Grid Connected Data Centre Projects in Great Britain

The 29 July 2026 "Curate - Demand Connections Reform" consultation (the Consultation) sits within the wider Demand Connections Reform programme, led jointly by Ofgem, the National Energy System Operator (NESO) and the Department for Energy Security and Net Zero (DESNZ). The three pillars of these reforms are as follows:

• Curate: reduce speculative projects in the grid connection queue and ensure only viable projects enter and proceed to connection.
• Plan: accommodate growth in the data centre sector alongside the wider energy system by prioritising strategic projects.  
• Connect: accelerate and expand physical grid connections, particularly through alternative connection models.

Ofgem reports in the Consultation that demand connection applications rose from 41 gigawatts (GW) to 125 GW between November 2024 and June 2025, with data centres accounting for around 73 GW of that total increase. This consists of 315 data centre projects with contracted capacities ranging from 1 megawatt (MW) to 1,500 MW, with 99% of the total data centre capacity in the demand queue comprised of projects above 40 MW. 

Under the grid connection regime for generation projects - the TMO4+ gated connections process - projects must generally satisfy both "ready" and "needed" criteria to obtain a "Gate 2" offer. To date, all demand connections (such as those for data centres) have been deemed “needed”, and had a lower threshold to be able to join the grid connection queue. Ofgem now consider that the relatively low cost of securing and retaining a queue position for a demand connection has incentivised speculative data centre projects to join and hold a place in the queue, and is placing significant stress on the network’s ability to plan – with the risk of inefficient planning and delivery falling on electricity bill payers. 

The introduction of a commitment fee, coupled with strengthened progression milestones akin to those introduced for generation projects, aims to give priority to those data centre projects with a genuine, demonstrable intention and capability to progress to energisation. 

For developers and investors considering new data centre projects, or already holding a place in the connections queue, the proposals mark a material shift in the cost, development risk and evidence burden attached to securing and retaining grid capacity. 

Who is caught?

The proposed fee would apply to data centre projects with a requested capacity of 40 MW or more and would capture both new and existing projects, whether connected at transmission level or distribution level (if they trigger a Transmission Entry Assessment). 

All demand projects would be required to self declare whether they qualify as a data centre. Penalties would be imposed for failure to comply with this requirement or for misrepresentation.

How much? 

Ofgem is proposing a fee of between 2.5% and 7.5% of average data centre capital expenditure, which is equivalent to roughly £237,500 to £712,500 per MW of requested capacity.  

For a 40 MW project (the minimum size of project to which the fees are proposed to apply), this could result in a commitment fee in the range of £9.5 million to £28.5 million. For larger projects, this could increase significantly. For the largest project of 1,500 MW reported to be in the queue, this would be in the range of £356.25 million - £1.07 billion.

Neither the fee level nor the 40 MW threshold is fixed at this stage. Further, Ofgem proposes a governance mechanism, embedded in NESO’s licence, allowing both to be reviewed and amended over time.

When does it need to be paid? 

The fee would be secured from grid offer acceptance and returned in full at energisation if the project proceeds as planned. It would be forfeited in circumstances where the project terminates after offer acceptance, reduces its requested capacity, or is found to have misrepresented its status as a data centre. 

Critically, given the interaction with the new proposed project milestones (see below) this commitment fee would be required to be paid potentially before full land and planning rights have been secured for the project. This therefore introduces a new, significant project risk before the point in time when a final financial investment decision would typically need to be taken for the project as a whole.  This will have a significant impact on developers’ investment decisions and timing considerations in respect of project development.   

Projects that are in the queue before these new requirements come into effect would have a period in which they would be permitted to self-terminate before the fee security is implemented, without triggering forfeiture of the fee.

How is it secured? 

The fee would be treated as a security under the existing CUSC securities framework, using the same accepted forms as other connection securities, including: 

(a)    a performance bond or letter of credit from a qualified bank;

(b)    a cash deposit into a bank account;

(c)    a performance bond from a qualified company.

As a second part of this major reform, the Consultation outlines plans for the introduction of evidence-based queue management milestones, which would operate alongside the existing NESO queue management milestones for generation projects, aiming to ensure that data centre projects which plan on utilising network capacity are making progress through their own development and construction in parallel with the connection arrangements. 

Satisfaction of the various milestones will depend on whether developers have elected for the Self-operation Pathway (for developers seeking to develop and operate the data centre themselves) or the Lease or Sale Pathway (for developers intending to lease or sell the data centre to a tenant, purchaser or third party), or a combination of both. 

Proposed additional data centre focused milestones and their associated evidence requirements

Milestone: M0.5.Dc: Pathway selection and non-binding compute offtaker evidence 

Timing: Within six months of signing a connection agreement 

Evidence required: Confirmation of pathway selection and non-binding compute offtaker evidence, such as a Heads of Terms or a Memorandum of Understanding or another customer contract (such as a Master Services Agreement). 

Milestone: M2.Dc: Long-lead procurement evidence

Timing: At the same time as the existing M2 milestone (Secured Statutory Consents & Planning Permission)

Evidence required: Proof of procurement of critical data centre equipment such as transformers, switchgear or other long-lead electrical equipment. This could be evidenced by a purchase invoice, procurement contract or supplier commitment. 

Milestone: M6.Dc: Financial and technical capability evidence 

Timing: At the same time as the existing M6 milestone (Agree Construction Plan)

Evidence required: Financial capability would be evidenced by: 

• a credit rating of at least BBB (Standard & Poor’s), BBB (Fitch), or Baa3 (Moody’s).
• credit support in respect of the project such as a bank letter of credit, cash deposit or another equivalent form of financial security.

Technical capability would be evidenced by: 

• an EN50600 certification including a design conformity certificate following assessment of the project against EN 50600-1 and EN 50600-2.
• TIA-942 Design Certification from a certification body licensed by the TIA.
• Uptime Institute Tier Certification of Design Documents, including evidence that the certification has been obtained at Tier I, II, III or IV.

Projects on the Lease or Sale Pathway would also need to provide binding compute offtaker evidence (the form of which remains subject to review). 

Scope and implementation

The queue management milestones are proposed to apply to projects with a 10MW rated IT load. This has been deliberately set lower than the threshold proposed as part of the commitment fee regime, to deter projects from bunching just below the 40MW threshold in an attempt to avoid scrutiny altogether. 

The milestone proposal would apply at both transmission and distribution levels. The CUSC framework will largely govern the implementation at transmission level while distribution level implementation will require amendments to contractual agreements, regulatory reporting requirements and queue management guidance.

Non-compliance with the milestones will likely have stringent sanctions which may result in removal from the connections queue. 

1.    A new risk profile. The commitment fee and the progression milestones represent a different risk profile for new data centre projects compared with the existing position. Ofgem's stated aim is to deter non-viable and speculative developers from entering, or remaining in, the connections queue. We expect that the actual impact will however be wider than eliminating speculative projects only; the introduction of a flat fee of this scale in a data centre’s development stage will also have negative impacts on those projects which may be viable in due course, but are not able to take that level of risk when (for example) there are still planning or land hurdles to overcome, which remain outside of the developer’s control. As a result, the proposed commitment fee could have a significant cooling impact on the UK’s modern industrial strategy.

2.    Review of portfolio strategy. Developers holding multiple queue positions, particularly "powered land banking" strategies, are an intended target of the proposal and should expect increased pressure to sell-off surplus projects to other developers or bring projects to energisation.

3.    Keeping self-declaration under active review. Projects should keep their data centre classification under review, particularly if their technology or use case changes. Captured projects will be expected to post additional security to cover any increase in fee levels if Ofgem or NESO adjust the commitment fee or the applicability threshold.