Failure Examples: How Britishvolt Built Capacity Before Securing Buyers
Britishvolt is usually remembered as a funding failure. The company ran short of money, missed the milestones attached to a government grant, and entered administration in January 2023 with most of its staff made redundant. That account is accurate as far as it goes, and it explains very little.
The more useful reading concerns sequencing. Britishvolt had identified a market that was genuinely expanding, and it had secured land, planning consent, raw material supply and conditional private commitments worth £1.7bn. What it had not secured was a customer contractually obliged to take volume. Its relationships with carmakers sat at memorandum of understanding level throughout, and the cells those carmakers were expected to buy were still in testing a year after construction began.
The company committed fixed capital against a demand signal read from policy, forecasts and investor appetite rather than from its own order book. Most of what followed is a consequence of that ordering.
What Britishvolt Set Out to Build
Power by Britishvolt Ltd was incorporated in December 2019. Within two years it had assembled one of the largest industrial propositions in the north east of England.
Everything rested on a single site at Cambois, near Blyth in Northumberland, on land formerly occupied by Blyth Power Station. The parameters were set out repeatedly and publicly:
A £3.8bn battery cell plant, which the company called a gigaplant
An initial 10GWh phase rising to roughly 30GWh across three phases by 2027
Output sufficient for more than 300,000 electric vehicles a year
Around 3,000 direct jobs on site and a further 5,000 across the supply chain
First cells leaving the production line by the end of 2023
There was no phased entry across smaller facilities and no proving plant carrying early volume while the main site was built. Capacity, jobs and capital sat on one plot of land.
Planning permission was granted in July 2021 and site works began shortly afterwards. The physical build was therefore under way before the product had been validated with a buyer and before the majority of the money required to finish it had been raised. The £3.8bn figure was a total project cost, not a sum held in the company's accounts.
The Difference Between Market Demand and Customer Commitment
Britishvolt's commercial position was not empty. By September 2022 the company had signed memoranda of understanding with five vehicle manufacturers, including Lotus and Aston Martin, and across those relationships had reached two joint development agreements and one pre-offtake agreement. Development cells had passed UN38.3 safety certification and were being distributed to seven customers for testing.
That is a pipeline with nothing closed at the end of it.
A memorandum of understanding sets out an intention to work together. It does not oblige either side to buy or supply anything, and the Aston Martin agreement was described at the time as falling short of a binding contract. A joint development agreement funds engineering work towards a specification. A pre-offtake agreement anticipates a future offtake agreement without constituting one. None of these instruments produces committed volume, and committed volume was the basis of the business case.
Why that gap was fatal rather than inconvenient
McKinsey's Battery 2030 analysis, published the day before Britishvolt entered administration, made two observations that sit awkwardly together. Globally announced nameplate capacity for lithium-ion cell factories already exceeded McKinsey's own demand projections. And battery cells are not traded on a free-floating spot market; they are sold through long-term supplier contracts.
Together those points describe a market with no safety net for an uncontracted producer:
A plant without long-term contracts has no channel through which to sell what it makes, because the channel is the contract
Announced capacity exceeding projected demand meant buyers had options and could wait
Buyers were comparing an unbuilt site against operating plants with delivery records
The signal that was being read
The demand evidence Britishvolt relied on was real but external. The 2030 prohibition on new petrol and diesel car sales in the United Kingdom, carmakers' own electrification targets, the Automotive Transformation Fund and the political framing of the project as a levelling-up opportunity all pointed in the same direction. None of them supplied a purchase commitment.
Why the Supply Chain Was Locked Before the Order Book
Britishvolt understood contractual commitment perfectly well. It pursued it in one direction.
On the supply side the company built genuine contractual structure. It held a cobalt offtake agreement with Glencore, which was also an investor. It signed with POSCO Chemical in June 2022 to secure cathode and anode materials. It reached an agreement with the Indonesian firm VKTR for nickel sulphate. It entered a battery recycling joint venture with Glencore anchored at an existing refining operation in Northfleet.
The approach was stated openly. Britishvolt's then chief executive described allocating space on the capitalisation table to raw materials producers, and argued that securing offtake with miners mattered as much as securing it with vehicle manufacturers. Relationships with carmakers were framed instead as joint exam questions, a phrase suggesting shared engineering enquiry rather than a purchase.
What that produced structurally
Suppliers had certainty of demand from Britishvolt
Investors had exposure to a plant with no committed revenue
Carmakers had optionality at no cost, able to walk away without penalty
Britishvolt carried the entire risk of the gap between the two ends
There is a defensible logic underneath it. Raw material availability was a widely discussed constraint in 2021 and 2022, prices were volatile, and a producer without secured inputs could find itself unable to manufacture at any price. Locking supply early was a real risk being managed.
It was also the second-order risk. A producer with secured inputs and no buyers has a warehouse. A producer with committed buyers and unsecured inputs has a procurement problem, which is difficult but negotiable, and a contract in hand with which to negotiate. Britishvolt resolved the tractable constraint and left the existential one open.
Capital Raised Against a Plan Rather Than a Pipeline
The funding announcements were substantial and, in their own terms, accurate. In January 2022 Britishvolt confirmed an in-principle grant offer of around £100m through the Automotive Transformation Fund, delivered by the Advanced Propulsion Centre. Days later, Tritax and abrdn committed £1.7bn towards the project.
Both were conditional, and the conditions were sequenced.
The government grant was contingent on Britishvolt meeting specified milestones, including construction progress and private sector investment commitments. The Department for Business, Energy and Industrial Strategy was explicit on this point after the collapse, describing support offered on the condition that key milestones were met. Funds were released against delivery. The private commitment, in turn, followed the grant offer and depended on the project proceeding as described.
Conditionality as external risk pricing
Two sets of experienced funders, one public and one institutional, looked at the same proposition and declined to release money against the plan alone. They attached delivery tests instead, and those tests were priced into the paperwork from the beginning:
Neither funder was exposed to the plan failing, only to it succeeding more slowly
The company carried the cost of reaching each milestone before receiving the money attached to it
Missing one milestone withheld capital needed to reach the next
Announced totals therefore described a ceiling, not a balance
Britishvolt reported the headline figures as secured funding. In substance they were options held by others, exercisable on Britishvolt's performance. The questions that would have distinguished the two are narrow ones: what proportion has actually been drawn, what triggers the remainder, who decides whether a trigger has been met, and what happens to the rest of the structure if one trigger is missed.
The Cost Base Arrived Before the Revenue
By the autumn of 2022 Britishvolt was running an organisation of roughly 300 people across three locations: the Cambois site, a headquarters function, and research and development activity in the Midlands. Following the collapse, EY confirmed 232 staff on the books at the point of administration, of whom 26 were retained to assist with the sale of the business and assets.
A plant of this type requires engineering, process, quality, procurement and commercial capability in place well ahead of first output. The difficulty was that the period between building the organisation and earning revenue kept extending.
A schedule that moved in one direction
Originally, first cells off the line by the end of 2023
Then 2024, following changes to the project
Then mid-2025, attributed to a factory redesign alongside inflation and rising interest rates
Site works were halted in August 2022, before any of those dates arrived
Each delay compounded. The organisation continued to run while the revenue date receded, so the capital required to reach first sale rose every time the schedule moved. The grant milestones were construction-linked, meaning delay simultaneously increased the cash requirement and postponed access to the cash available.
By November 2022 the company had secured short-term funding and its staff had agreed a temporary pay cut while longer-term investment was sought. At that point the cost base had stopped being an investment in future capacity.
When Product Validation Runs Behind Construction
Britishvolt's cells passed UN38.3 safety certification in September 2022, clearing them for transport, and development cells then went to seven customers for testing.
That was fourteen months after site works began at Cambois. Testing is where a buyer establishes whether cells meet its requirements on energy density, power delivery, charging behaviour, durability and cost. It is the point at which a memorandum of understanding either converts into an order or does not.
Britishvolt reached it after committing to the site, the headcount and the build schedule, and while requiring further capital to complete a plant whose output no buyer had yet accepted. The validation that would have justified those commitments was still in progress when the money ran out.
The sequencing had a second effect on negotiating position. A supplier holding test results and a qualified product negotiates from evidence. A supplier still in testing, visibly short of funding and behind schedule, is asking buyers to commit volume to a plant that might not be finished. By late 2022 Britishvolt was in the second position, and the carmakers it needed had no reason to move.
How the Position Unwound
The sequence from halt to liquidation took under two years.
Site works at Cambois stopped in August 2022. Orral Nadjari, who co-founded the company, had left the previous month, with Graham Hoare taking over. In November the company secured short-term funding and its staff accepted a temporary pay cut while longer-term investment was sought. In early January 2023 Britishvolt confirmed it was in discussions with a consortium of investors over a potential majority sale.
Those talks failed. On 17 January 2023 the directors appointed Dan Hurd, Joanne Robinson and Alan Hudson of EY-Parthenon as joint administrators. EY attributed the insolvency to insufficient equity investment for the company's research and for the development of its sites in the Midlands and the north east. Of 232 staff, 26 were retained to assist the sale.
What the assets were worth
The majority of the business and assets sold to Recharge Production UK Limited on 26 February 2023 for £8.57m, payable in instalments
The purchaser failed to pay the final instalment and remained in default a year later, with administrators reopening discussions with other parties
Ashtead wrote off a £35m investment; NG Bailey, an investor and contractor, was owed more than £2m
Administrators ceased to act on 28 October 2024 and the company entered liquidation
Set against a £3.8bn project and £1.7bn of announced private commitment, £8.57m is the recoverable value of intellectual property, research assets and equipment belonging to a company that had never sold a cell commercially. The land sat in a separate entity, over which a secured lender had appointed receivers in November 2022, two months before administration.
What the Site's Second Life Says About the Asset
The 235-acre site was sold in 2024 to Blackstone, whose subsidiary QTS is developing a data centre campus of up to 540,000 square metres. Northumberland County Council granted outline permission in March 2025 and approved the first phase that December. Enabling works began in October 2025. The investment is put at up to £10bn, with around 1,200 long-term construction jobs and hundreds of operational roles.
The site retained its value. What made Cambois attractive to Britishvolt was what made it attractive to QTS: a decommissioned power station with heavy grid capacity, coastal access and consented industrial land. Ten buildings drawing 720MW require the same underlying infrastructure a gigaplant would have used.
Britishvolt secured a genuinely scarce asset and could not build a business on top of it. The land found a buyer within fifteen months of administration. The battery operation, its intellectual property and its research assets sold for £8.57m and the purchaser then defaulted.
What Expansion Teams Can Take From This
Britishvolt was a manufacturing business and the capital scale involved has few parallels elsewhere. The sequencing questions transfer regardless of sector.
Separate market evidence from customer commitment. Market sizing, regulatory direction, competitor activity and inbound interest establish that a market exists. A signed agreement establishes that a named buyer will transact, at a stated volume, at a price, on a date. The first justifies investigation. The second justifies fixed cost.
Establish how the market transacts before sizing it. Where access runs through long-term negotiated contracts held by a small number of buyers, capacity without those contracts has no route to revenue. Growth in a market says little about whether a new entrant can reach it.
Take the hardest commitment first. Supplier-side agreements are easier to obtain because the counterparty gains from granting them. Progress on that side is a weak proxy for commercial progress, and it accumulates obligations without producing income.
Read funders' conditions as diagnosis. Milestones attached to capital show where experienced funders judge a plan to be weakest. Announced totals describe a ceiling. The drawn figure describes the position.
Cost the delay rather than the plan. The question is what happens to the funding requirement if first revenue moves by twelve months, and then by twelve more. Overhead committed early turns a timing problem into a solvency problem.
Validate before the commitment that validation is meant to justify. A supplier holding test results negotiates from evidence. A supplier still in testing, behind schedule and visibly short of funding, is asking buyers to take the risk instead.
Check whether the asset survives the business. Cambois held value independent of the venture built on it. Most expansion commitments do not, and the recoverable value of a failed market entry is usually a fraction of what went into it.
Applied to Britishvolt in 2021, these questions all point at the same missing item. The company had secured land, consent, inputs, conditional capital and political support, and had not secured a buyer obliged to purchase anything. The cost of that gap was still recoverable at the point it could have been identified.
Frequently Asked Questions
Why did Britishvolt fail?
The immediate cause was insufficient equity investment, the reason EY gave on appointment as administrator in January 2023. The underlying cause was sequencing. Britishvolt committed to a £3.8bn site, a 300-person organisation and a construction programme before securing any customer contractually obliged to buy its cells. Funding ran out while the product was still in customer testing and the plant was unfinished.
Did Britishvolt ever have confirmed customers?
No binding volume commitments. By September 2022 it had memoranda of understanding with five vehicle manufacturers, including Lotus and Aston Martin, plus two joint development agreements and one pre-offtake agreement. None of those instruments obliged a buyer to purchase anything. The Aston Martin agreement was described at the time as falling short of a binding contract.
How much government money did Britishvolt actually receive?
Less than the headline figure suggests. The grant offer through the Automotive Transformation Fund was around £100m, but release was conditional on milestones including construction progress and private investment commitments. Those milestones were not met. The government confirmed after the collapse that support had been offered on that condition, which is why the announced sum and the amount available to the company differed.
Who bought Britishvolt and what happened to the Blyth site?
They went separately. The business and assets sold to Recharge Production UK Limited in February 2023 for £8.57m in instalments, and the purchaser later defaulted. The land sat in a different entity and was sold to Blackstone in 2024. Its subsidiary QTS is building a data centre campus at Cambois, with enabling works from October 2025 and up to £10bn of investment.
Is Britishvolt still trading?
No. Administrators ceased to act on 28 October 2024 and the company entered liquidation.
What can other companies learn from the Britishvolt collapse?
Chiefly that market growth and market access are separate questions. Battery demand was rising steeply, but cells are sold through long-term negotiated contracts rather than an open market, so a plant without those contracts had no route to revenue. The wider lesson concerns order of commitment: validate the product and secure the buyer before committing the fixed cost built to serve them.
How Metheus Can Help
We work with technology, software and fintech businesses on the sequencing decisions that sit behind market entry and expansion. That usually means separating evidence that a market exists from evidence that a named buyer will transact, establishing how a market actually transacts before committing to serve it, and identifying which commitments can safely be made early and which should wait for a signed agreement.
We test demand assumptions against the way buyers in a given market are structured and what they are contractually willing to sign, then build the entry plan around that rather than around a forecast. Where a plan depends on conditional capital, we help clients read those conditions as an assessment of the plan.
References:
McKinsey, Battery 2030: Resilient, Sustainable, and Circular
EV Specifications, Britishvolt Passes Battery Safety Tests UN38.3
Fastmarkets, Raw Materials the Priority as Supply Constraints Intensify, Britishvolt CEO Says
Construction News, Britishvolt Administrators Look for New Buyer After Deal Stalls
edie, Government Funnels £100m Into Britishvolt's Battery Gigafactory