What the qualification window cannot test

Somewhere inside NESO at the moment, a small number of people are working through the Allocation Round 8 applications and deciding which projects are allowed to bid at all. The qualification assessment window opened on 10 August and closes on 16 September, and the following morning DESNZ publishes the Contract Budget Notice and issues the Notice of Auction. Between now and the results in late November, this is the part of the round that shapes most of the outcome, and it is attracting a fraction of the attention that the clearing prices will get in December.

Worth being clear about what has actually changed, because the headline numbers have not. The Administrative Strike Prices for AR8 are the same as AR7: £113/MWh for fixed-bottom offshore wind, £271/MWh for floating, £92/MWh for onshore wind and £75/MWh for solar photovoltaic, all expressed in 2024 prices. Government held those ceilings deliberately, on the stated grounds that stability between rounds supports the competitive price outcome it is after. What tightened instead was the door.

A project now needs Gate 2 connection status to enter, with one narrow transitional exception for projects holding a Gate 1 Connection Point and Capacity Reservation. The reason given is that a Gate 1 agreement no longer provides enough certainty that a project will progress to Gate 2 and connect within the delivery timeframe. Capacity surrendered under an earlier CfD is now permanently barred from this round and future ones. And the default bid has gone, so an applicant that qualifies but fails to submit a bid is treated as withdrawn rather than being handed the ceiling price.

Read those changes together rather than one at a time and the round is asking a different question from its predecessor. AR7 asked what a project would cost. AR8 asks whether it can be built. On the face of it that is progress, and I think it genuinely is progress - the framework has learned something from watching capacity get won and then handed back.

But this is where it gets more complicated. Every one of those tests is documentary: a signed connection agreement, a Gate 2 to Whole Queue notification from NESO, land rights at a stage the framework will accept. Each is a real thing and each is worth testing, and each of them proves that a project has secured something rather than that the organisation holding it can build anything.

Nothing in the AR8 framework asks whether the joint venture that owns the consent has settled how it will make decisions once the money is committed. There is no test of whether the owner, the delivery partner and the tier one contractors share an understanding of who does what in the first six months after award, still less of whether the three of them have ever put that understanding under any real pressure. The framework does not ask, and I am not sure what a workable version of the question would even look like inside an auction.

I should be clear that I am not arguing DESNZ ought to be testing any of it. The CfD is an allocation mechanism, and asking it to assess how well a consortium is likely to work together would be asking it to do something it was never designed to do, with tools it does not have. Fair enough. The point is narrower and, to my mind, more awkward. Nobody is testing it, and each round of tightening makes the omission more visible rather than less. The framework is getting steadily better at filtering out projects that cannot be built for documentary reasons. It is no better than it was three rounds ago at anticipating the projects that will be built late and expensively for organisational ones.

Advance's own work on this, published as Six Months is Too Late, argues that the window in which mobilisation problems can be fixed cheaply opens the day the contract is awarded and closes considerably earlier than most programmes assume. That is our professional judgement rather than a finding anyone independent has verified, and it should be read as such. What is not our inference is the direction of travel in the auction design, which is now explicitly about deliverability. Government has said so itself, in the reasoning attached to both the Gate 1 exclusion and the permanent restriction on surrendered capacity.

There is a version of this argument that overstates itself and it is worth avoiding. Plenty of consortia sort this out quietly and without any help, and the sector has delivered a great deal of offshore wind while people like me were worrying about whether it was organisationally ready to. The uncomfortable consistency sits in the distribution rather than the average. The programmes that struggle tend to struggle for the same handful of reasons, and those reasons are visible early to anyone who goes looking for them.

On 17 September the Contract Budget Notice lands and the conversation will move immediately to budget, pots and what the clearing prices are likely to do. Reasonable enough, since that is the commercially important number and everyone in the round has been waiting for it. The question I would put alongside it is which of the qualified projects have used the fourteen months since AR7 closed to work out how they will actually operate together once the contract is signed, because those are the ones still on programme in 2029.

The sealed bid window opens on 1 October, which happens to be the first morning of GRID at ExCeL. Ten working days in which every qualified developer will be thinking very hard about price. Whether any of them are thinking as hard about the first six months after award is what I intend to be asking on the floor

SOURCES:

Contracts for Difference Allocation Round 8 statutory notices, DESNZ, published 23 April 2026 and last updated 20 July 2026, including the AR8 Pot and Price Notice, which confirms Administrative Strike Prices maintained at AR7 levels in 2024 prices. AR8 Contract Allocation Framework, DESNZ, final version published 6 July 2026, for the Gate 2 eligibility requirement, the permanent restriction on surrendered capacity and the removal of default bids. AR8 indicative timetable as published by DESNZ for the qualification assessment window, the Notice of Auction date and the sealed bid window. Six Months is Too Late, The Advance Consultancy.