The Queue Is the Cost: Pricing Grid-Connection Delays across New Assets

1 Define the connection service and the clock

This paper proposes a capital-allocation method for projects whose commercial start depends on electricity-network access. Its unit of analysis is a specified service at a specified site: capacity, direction of power flow, operating restrictions and the date on which that service becomes usable. A renewable generator, an industrial plant and a data centre can share a network constraint while requiring different connection arrangements. The proposed method therefore records the applicable process before comparing delay exposure. No private connection agreement, equipment quotation or actual project cash forecast has been supplied.

The US Department of Energy's April 2024 interconnection roadmap focuses on bulk transmission connections. It explains that large-load processes generally sit outside the traditional generation, storage and transmission queues it examines. Its request-to-agreement timing target also excludes the subsequent interval to commercial operation. Broader supply-chain, permitting and siting issues were outside its principal scope. [4] Those boundaries support a basic diligence question: which interval does a claimed acceleration actually shorten?

The proposed project schedule distinguishes application validation, studies, an acceptable offer, satisfaction of conditions, network works, energisation and the customer's commercial ramp. Each milestone needs an owner, documentary basis and dependency. If two activities can proceed concurrently, the schedule should show that relationship. Adding their durations would overstate the delay. If an activity starts only after another finishes, record the dependency and any waiting period explicitly. A headline connection date should have a stated meaning throughout the model.

For an illustrative renewable development, the revenue start might require both export capability and completed commissioning. For an illustrative industrial facility, initial imported power might support testing while full production requires a larger supply. For an illustrative data centre, the first available capacity block might support a limited customer deployment. These are proposed scenario distinctions. The actual contracts and technical design must establish whether such staging is available and whether customers accept it.

Define the baseline before calculating loss. It should contain the same service specification, operating end date and capital scope as the delayed case unless the purpose of the comparison is to change them. Separate a planning estimate from an accepted contractual date, and record remedies only when the applicable agreement supports them. The resulting analysis asks what each delay changes: the start of receipts, the length of earning life, the amount of capital committed or the rights available. This creates a common financial language across asset types while preserving their different physical and regulatory requirements.

Table 1. Define the service before valuing a connection

Proposed comparison framework. No actual capacity, reliability standard, supply permission or customer receipt is established.

2 Use queue evidence with an explicit denominator

Berkeley Lab's June 2026 summary covers US generation and storage requests through the end of 2025. It reports approximately 8,200 active projects, comprising 1,312 GW of generation and about 749 GW of storage. For regions with available data, projects built in 2025 had a median request-to-commercial-operation interval exceeding five years. The source also states that the full effect of interconnection reforms remains too early to assess. [5] These are dated generator and storage observations. They provide no measured waiting-time distribution for a new industrial or data-centre load application.

The proposed use of historical evidence begins with a cohort definition. Record the geography, network level, technology, entry period and endpoint. A statistic for completed projects describes that completed group. An active queue contains applications whose final outcomes are unresolved. A withdrawn project may have incurred expenditure without ever earning operating revenue. A useful investment comparison should keep those outcomes visible and state which group each statistic describes.

The analytical challenge is especially important when management wants a single delay probability. This paper recommends presenting dated scenarios before assigning probabilities. If management supplies probabilities, identify their author, rationale and update date. A national median for completed plants supplies neither the probability of an individual site's connection nor the distribution of its remaining wait. A queue position also needs interpretation under the applicable process; the model should not assume that all applications ahead consume the same resources or will remain active.

Build an evidence table that can be updated without rewriting the investment case. Each entry should connect a forecast date with its source, the required preceding event and the next point at which the forecast can be tested. For example, a study-completion target could be checked against the study agreement and current operator correspondence. An equipment-delivery date could be checked against the supplier's acknowledged order and delivery conditions. Where the evidence is missing, the case should preserve an explicit assumption and a sensitivity around it.

For portfolio analysis, avoid treating several sites as independent solely because they have separate project companies. A proposed common-cause scenario can place all projects relying on the same substation works or equipment order under one delay event. A separate scenario can test an individual site's readiness failure. Neither scenario requires an invented empirical correlation. The investment committee can then see whether diversification by asset label also diversifies the actual connection dependency, and how a shared delay changes the timing of portfolio cash requirements.

3 Translate US reforms into project-level commitments

FERC's official explanation of Order 2023 describes generator cluster studies, stronger financial-readiness and site-control requirements, withdrawal penalties and procedures for affected systems. It states that transmission-provider implementation follows approval of compliance filings and cautions that the explainer is not a legal document. [1] The paper uses these features to identify questions for project diligence. The actual provider's approved tariff and agreements remain necessary to establish applicable obligations, costs and deadlines.

The official Order 2023-A explainer identifies a change allowing surety bonds or other reasonably acceptable financial security alongside cash and irrevocable letters of credit for specified deposits. [2] The proposed financial analysis should therefore distinguish the required face amount from the cash that the chosen instrument ties up. Obtain evidence for collateral, fees, expiry, renewal and draw conditions. A nominally noncash instrument can still consume a sponsor's facility or require cash collateral; the amount belongs in the model only when terms are supplied or explicitly assumed.

Large-load reform has a separate procedural history. On 18 June 2026, FERC announced tailored show-cause orders for six jurisdictional regional grid operators, asking them to justify or reform large-load tariff arrangements. Its stated categories included study processes, cost transparency, co-location, flexible services and generation serving nearby loads. [3] This is evidence of a dated regulatory action. The paper has not verified subsequent regional determinations and derives no uniform current connection right from that announcement.

For an investment decision, map each rule-related payment to its economic treatment. Study expenditure consumed by completed work belongs among incurred costs. Refundable security needs a release assumption and a cash-availability test. A withdrawal amount needs the triggering event and calculation mechanism from the actual documents. Proposed network-upgrade contributions need their estimate version and any circumstances that could change the allocation. The reviewer should identify which amounts are commitments, estimates or contested allocations.

A hypothetical withdrawal decision illustrates the distinction. Compare future spending required to continue with the future receipts and rights available under continuation. Then compare the cost of exit, including any supported security loss or contractual payment. Historical expenditure should remain in the funding reconciliation, while the forward choice should identify the amounts still avoidable. If withdrawing could affect another project in the same cluster, request the applicable assessment and allocation evidence. This paper assigns no universal withdrawal charge or automatic recovery from another customer. The proposed output is an expenditure decision with documented exposure, including the next point at which continuation can be reconsidered.

4 Separate British queue reform from delivery and demand proposals

NESO's retained evidence guidance describes Gate 2 as requiring readiness and strategic alignment. Gate 1 projects receive no confirmed connection date under that guidance. The page identifies the evidence window as closed and retains resources for required resubmissions. [6] Those descriptions concern process status. For a particular investment, request the actual outcome, offer and outstanding conditions before using a connection date in the committed case.

The published NESO timeline concerns offer issuance. Its phased windows include Gate 2 Phase 2 distribution offers between mid-October 2026 and mid-March 2027. The February 2026 statement explains revisions arising from technical matters, background assumptions and construction planning; the page also records June regulatory authority for the timelines. [7] It does not certify that every offer was delivered or that a project can energise within the offer window. Ofgem's 21 April 2026 update separately acknowledged that many protected projects with 2026 or 2027 connection dates would not receive their original dates. [8]

Demand-side policy also requires a dated status check. Ofgem's 29 July 2026 consultation proposes a data-centre commitment fee and evidence milestones addressing credible end-users, long-lead electrical equipment and development capability. At this review, its official page records the consultation as closed and awaiting a decision. [9] On 7 October 2026, Ofgem opened a separate consultation on expanding self-build and transfer options for sole-use transmission assets serving demand, generation and storage customers. [10] Both are proposals at the inspected stage; this paper models no entitlement arising from their adoption.

The proposed British project review should maintain three linked records: regulatory status, the customer's connection documents and the network delivery programme. A policy consultation belongs in the first record. A revised offer belongs in the second. Construction progress belongs in the third. Movement in one record should trigger a review of the others without silently treating them as equivalent evidence. If a proposed reform could change future security requirements, show that as a policy sensitivity with an explicit decision trigger.

For site selection, compare the service actually offered, its conditions and its dependency on wider works. A favourable queue classification can inform that assessment but leaves questions about cost, construction and customer readiness. The proposed decision memorandum should identify what can be committed immediately and what remains contingent on a further offer or engineering result. This makes the capital decision responsive to reform while retaining the specific delivery evidence required for an irreversible order.

5 Price local power constraints in the customer's operating model

The IEA's 2025 Energy and AI analysis examined prospective data-centre locations, congestion and connection conditions. Its scenarios indicated that grid constraints could delay around 20% of global data-centre capacity planned for construction by 2030. [11] This is a modelled prospective exposure. It cannot be used as an observed failure rate or applied automatically as a probability to each site. The IEA's 16 April 2026 update identifies continuing equipment supply-chain constraints and planning pressures affecting grid connections and other approvals. [12] That broad diagnosis supplies no quoted delivery time for an individual project.

The proposed customer model starts with the usable power profile. Record the initial capacity, subsequent increments, operating restrictions and any supported interruption arrangements. Translate that profile into the service the business can actually deliver. For a data centre, ask the operating team which deployment stages and customer obligations fit the available supply. For an industrial facility, ask which production steps can operate and what an interruption would require. For a renewable project, translate permitted export and commissioning assumptions into the modeled energy receipts. All technical relationships require project-specific validation.

An alternative-power scenario should carry its own complete scope. A proposed temporary supply needs evidence of permissions, equipment, fuel or energy procurement, operating costs and the customer's acceptance of the resulting service. A proposed flexible connection needs a defined restriction pattern or explicit scenarios where that pattern is unknown. Price the loss of usable service alongside any earlier start. A lower connection expenditure or earlier energisation date can coexist with a lower operating margin or a narrower commercial product.

The capital schedule should reflect those choices. A staged project may defer equipment orders and customer fit-out, incur extra mobilisation costs or sacrifice purchasing discounts. Each effect needs an assumption or quotation. Separate capital that can be moved without material cost from orders with cancellation exposure, warranties beginning before use or storage requirements. The proposed comparison should also show working capital and customer payments by date, including any contractual acceptance conditions that must precede collection.

The resulting value comparison should state the service boundary. If one option delivers fewer production hours, a different reliability commitment or less capacity, identify the adjustment instead of comparing headline project returns alone. Test the largest adverse change that the preferred option can absorb before another option becomes financially preferable. This break-even exercise can identify the engineering or commercial evidence worth obtaining next. It does not establish an observed market price for accelerated connection or a universal premium for a power-ready site.

6 Allocate capital against dated, reversible choices

The following decision framework is proposed analytical methodology. It values delay through a dated difference between a specified baseline and a specified alternative. Include operating cash moved or lost, changed capital timing, additional holding expenditure, contractual payments supported by evidence and the eventual release or consumption of security. Use consistent tax, currency and inflation assumptions. Identify whether the final operating date stays fixed or moves with the start; the answer changes how many earning periods survive.

Keep project value and funding availability in separate reconciliations. A postponed investment can reduce its present cost while increasing fees or extending a cash deficit elsewhere. Interest during construction belongs in the financing account under the chosen terms. If an unlevered project discount rate already incorporates financing effects, adding borrowing interest to the same project cash account needs a stated methodological justification. The reviewer should reconcile the chosen valuation convention with the sponsor's actual dated sources and uses.

The committee should then compare specific actions: continue the existing route, defer an order, procure a permitted alternative, relocate or withdraw. Each action needs a next commitment, its cancellation exposure and the evidence expected before the following commitment. An illustrative equipment reservation might preserve a delivery slot while allowing a later configuration choice; its value depends on the actual reservation terms and alternative schedule. A proposed self-build route should remain conditional until its legal availability, technical scope and adoption arrangements are established.

For a capital-constrained portfolio, rank the next funded decisions as well as the whole-project returns. A small study can resolve a material upgrade-cost uncertainty. A large irreversible order may require an accepted connection scope and a usable financing commitment. Record the information each expenditure is intended to obtain, whether it can benefit another option, and how much time remains to act on the result. Where probability estimates are unavailable, publish scenario and break-even results with that limitation intact.

Close the review with an executable authorisation: amount, recipient, permitted purpose, evidence conditions and reassessment date. Retain a dated downside showing what happens if the next network milestone fails or slips, including the funding needed to preserve, pause or exit the project. The proposed conclusion is a conditional capital decision supported by the current service specification and delivery evidence. National queue totals, regulatory announcements and global scenarios provide context; the investment requires a traceable connection between the site's actual obligations, its usable power date and the cash committed before that date.

7 Price a power-ready acquisition without buying the wrong service

The proposed valuation method can also be used when a purchaser compares a connection-ready site with one still awaiting network access. The object being acquired should be stated precisely: shares, land, equipment, a contractual position or a combination. The connection reviewer should establish which obligations and permissions can continue after the proposed transaction. The economic model then values the service that survives those conditions. This paper verifies no transferable connection right, executed acquisition or observed market premium.

An acquisition comparison should use the same commercial capacity, operating end date and customer boundary as the development alternative. A building that can energise earlier may still require expenditure to support the purchaser's intended load or operating commitment. Include that conversion expenditure, the assumed commissioning period and any relevant restriction on service. A headline electrical rating does not establish the operating receipts assumed in the acquisition case. The investment committee should require the engineering and commercial teams to reconcile that link before negotiating a price based on avoided delay.

The proposed maximum premium is a dated value difference after the costs of acquiring and converting the ready asset have been accounted for. Compute the difference between the chosen acquisition cash-flow case and the continuing development case before inserting the purchase premium. The present value of that difference is the model's available price headroom at the comparison date. If the premium is payable later, translate the headroom to that payment date using the declared discount convention. Transaction expenses and continuing liabilities should enter the calculation once, with their due dates visible.

This calculation supports a conditional negotiation position. It supplies no comparable-sale evidence and establishes no market value. A purchaser should test the headroom under the same delay, customer-ramp and operating-life sensitivities used for the development case. If the acquisition depends on a consent or third-party condition, retain that condition in its own timing scenario. A price adjustment or contingent payment can be proposed around evidenced capacity or acceptance events, subject to legal review and counterparty agreement. The board can then see which benefit is paid for immediately and which remains conditional on delivery.

8 Use one dated cash account for the capital decision

The proposed financial account starts at the date on which management can still change the next commitment. Identify cash already spent, liabilities already incurred and expenditure still avoidable. A historical study cost remains part of the project's complete record. The forward decision should use the future cash consequences of the alternatives available at the review date. When reporting the value of the original investment alongside a later continuation decision, label the dates and perimeters separately so that the different questions remain understandable.

Within the declared model, each annual operating amount represents cash available for debt service after assumed operating costs, tax, working-capital effects and ordinary lifecycle expenditure. Those components are embedded in the fictional inputs and are not independently forecast. A project-specific replacement should build them from dated contractual and operating evidence. Closure spending remains a separate cash event. The model should also identify any extraordinary equipment replacement or customer-conversion expenditure that changes between the alternatives. A simplified annual input should not conceal a cost that is specific to the preferred route.

The financing account should reconcile draw dates, capitalized interest, cash interest, principal and sponsor contributions. Construction interest is shown as debt growth before repayment begins in the illustrative financing cases. It is excluded from the unlevered project account. The final construction draw occurs at the assumed completion date; the first operating year's interest enters the first debt-service payment. These conventions require consistent use across every delay. Changes in reserve balances and financing flows cannot be counted as additional operating receipts in a coverage calculation.

The proposed audit trail has three checks. First, the unlevered value difference should equal the discounted sum of the dated cash differences. Second, opening debt plus allowed capitalized interest and draws, less principal repayment, should equal closing debt. Third, each dated funding requirement should be matched to a usable source or a separately identified gap. A satisfactory whole-project NPV leaves the third question open whenever funding conditions delay access to money. The investment recommendation should therefore show both the selected value case and the cash commitments that must be met before that value can be earned.

Table 2. Keep valuation and commitment evidence on the same calendar

Author-proposed decision record. Analytical value establishes no connection entitlement, financing commitment or completed approval.

9 Put the connection event on the investment calendar

The author-created model compares renewable export, industrial demand and data-centre demand investments using fictional USD millions. The decision date, t0, precedes the proposed investment. There is no existing debt, previously spent capital or inherited cash balance in the operating valuation. The three projects have different budgets and operating cash amounts; their absolute values do not rank real asset classes. Each operating route assumes that the required capacity, permissions and commercial arrangements become available on the stated date. The calculation establishes no connection offer, supply entitlement, engineering performance or financing commitment.

Total capital spending is 92 for renewable export, 145 for industrial demand and 210 for data-centre demand. Each budget has three separately dated parts. Development at t0 is 8, 12 and 18 respectively. Equipment spending at year 1 is 52, 83 and 112. Completion spending is 32, 50 and 80 at year 1 plus the connection delay. When delay is zero, equipment and completion spending occur at the same year-end. The model retains early equipment expenditure during delay; changing procurement commitments to defer equipment would require a different cash-flow schedule.

The first full operating receipt occurs at year 2 without delay. Annual CFADS is assumed to be 17, 29 and 44 respectively, received at year-end. These amounts include ordinary operating costs, ordinary energy costs for the demand projects, baseline tax and lifecycle cash. They exclude financing transactions and the separately stated holding, completion and closure expenditures. Demand CFADS measures the incremental business cash enabled by electricity access; it represents neither a connection fee nor electricity-export revenue. No production ramp, commodity-price change, tariff escalation or inflation is inferred.

Two end-date conventions are calculated. A fixed final date stops receipts in year 17, so a delay of one, two or three years removes that number of operating receipts. A shifted-life case moves the final date to year 17 plus the delay and preserves sixteen receipts. Closure payments of 9, 14 and 22 occur at the respective final dates. During each delayed year before operation, holding costs of 1.4, 2.8 and 4.2 are paid. Discount rates of 7%, 9% and 11% are analytical inputs, with 9% used for the principal comparisons. No empirical probabilities or probability-weighted expected values are assigned.

For each route, NPV at t0 equals the sum of CF_t / (1 + r)^t over all dated cash events, where CF_t is the signed incremental project cash flow at year t and r is the declared annual discount rate. Spending enters with a negative sign and operating receipts with a positive sign. A t0 payment has a discount factor of one. The calculation includes the route-specific completion, holding and closure payments, while debt draws, interest and reserve movements remain in the separate financing account.

Table 3. Declared fictional asset inputs

USD millions. Base receipt years 2-17; baseline operating costs, tax and lifecycle cash included in CFADS; financing excluded.

Table 4. Capex stage timing

All prospective at t0. When d=0, equipment and completion occur together at t1.

10 Price a lost operating year separately from a shifted life

At the 9% rate and without delay, incremental project NPVs are 42.501524 for renewable export, 83.906432 for industrial demand and 136.322409 for data-centre demand. With two years of delay and a fixed year-17 end, those values fall to 17.454363, 39.847959 and 70.154006. With three years of delay, they fall to 6.459604, 20.507951 and 41.108573. These reductions combine lost early operating receipts, holding expenditure and the beneficial discounting of deferred completion spending. Multiplying annual CFADS by the delay would omit both the timing of the cash and the staged investment response.

For renewable export, the two-year fixed-end reduction is 25.047161. It can be independently reconstructed by discounting completion spending of 32 from year 1 to year 3, then deducting the two lost receipts of 17 and two holding payments of 1.4 in years 2 and 3. Development and equipment spending remain unchanged. Closure still occurs in year 17, so it contributes nothing to this particular change. This bridge identifies the economic consequences of the specified delay without assuming that every cost moves with energisation. The same identity is checked separately for all three projects and each positive delay.

Preserving the full operating life produces different results. With a two-year delay, shifted-life NPVs are 24.693833, 52.148147 and 88.844114. The difference from the corresponding fixed-end case consists of receipts in years 18 and 19 and the discounted benefit of moving closure from year 17 to year 19. It therefore requires evidence that the relevant site, equipment, customer and operating arrangements can support the longer period. A delayed connection date alone establishes none of those extensions. The model keeps both cases visible so that a term restriction cannot disappear inside a generic terminal-value assumption.

The zero-value operating-cash thresholds provide another decision test. With a two-year delay and fixed year-17 end, annual CFADS must be 14.096909 for renewable export, 22.372301 for industrial demand and 32.331657 for data-centre demand to produce zero NPV at 9%. Under the shifted-life convention, the thresholds become 13.152902, 20.875743 and 30.158809. These are algebraic break-even amounts with the specified cost schedules retained. They establish no achievable tariff or customer margin. A separate holding-cost grid multiplies standby expenditure by one, two and three while preserving the connection and operating dates, making that isolated cost sensitivity reproducible.

The delay cases also assume that commissioning, customer acceptance and collection coincide with the first modelled operating year. A physical energisation milestone followed by testing, restricted output or unpaid invoices would need separate dates and receipts. Annual aggregation can conceal an intra-year funding deficit even when the displayed annual balance is sufficient. The present calculation therefore cannot certify monthly liquidity or working-capital adequacy. A transaction team would need to reconcile supplier payment terms, the energisation programme, acceptance tests and collectible operating receipts before converting an annual delay case into a usable draw schedule.

Table 5. Fixed-final-date connection delay

Hypothetical USD millions; holding cost and deferred completion capex included.

Table 6. Shifted-life connection delay

Hypothetical USD millions; holding cost and deferred completion capex included.

Table 7. Value of preserving the full operating life

Same delay and costs; shifted life retains 16 receipts and postpones closure.

Table 8. Discount-rate test with two-year delay

Fixed final year 17; rates are author assumptions.

Table 9. Operating cash needed for zero project NPV

9% discounting; two-year delay; each stated investment/closure schedule held fixed.

Table 10. Standby-cost sensitivity

Two-year delay and fixed final 17; only annual holding cash changes.

Author calculations in fictional USD millions at the assumed 9% discount rate. A fixed final date and a shifted operating life require separate evidence of the available term.

11 Compare conditional access routes within the same decision perimeter

The route comparison uses t0 before investment, a two-year full-connection delay and a fixed year-17 operating end. Waiting follows the original staged expenditure and receives its first operating cash in year 4. Phased access assumes that limited service can generate 35% of full CFADS in years 2 and 3, followed by full CFADS from year 4. Additional year-1 spending is 7 for renewable export, 11 for industrial demand and 16 for data-centre demand. Holding costs during the partial-access years fall to 50% of the waiting assumption. No incremental terminal cost or equipment recovery is included for that additional investment.

Under those assumptions, phased access produces NPVs of 21.764537, 48.396339 and 83.717718. Its gains over waiting are 4.310174, 8.548380 and 13.563712 respectively. The gain equals discounted early operating receipts and holding-cost savings less the additional investment. It depends on the ability of a limited connection to support the particular operating duty. The 35% factor is a fictional input. It should not be interpreted as evidence that a percentage of contracted electrical capacity yields the same percentage of export, production or computing cash flow.

Conditional alternative supply is calculated only for the two demand assets. No equivalent supply route is invented for renewable electricity export. Industrial demand spends 9 in year 1, earns annual early CFADS of 15.5 in years 2 and 3 after an incremental supply premium of 13.5, and receives assumed temporary-equipment recovery of 2 in year 3. Data-centre demand spends 14, earns early CFADS of 23 after a premium of 21, and recovers 3. Holding costs are 25% of the waiting amounts during those two years. Ordinary energy costs remain inside the original CFADS assumptions; the premiums represent additional costs, avoiding duplicate subtraction of the baseline bill.

These alternative-supply NPVs are 61.539443 and 101.829050. They assume that power quality, reliability, permissions and the required operating output can be achieved; none is verified. Relocation instead increases every original capital stage by 12%, adds t0 diligence of 5, 8 or 12, receives operating cash from year 3 through year 17, and increases closure spending by 3. Its NPVs are 13.828609, 36.607654 and 66.549906. Immediate exit costs 2.4, 3.6 or 5.4 and avoids the proposed investment. Each comparison is prospective and mutually exclusive. These values do not authorize relocation after the original equipment is already purchased, and the model supplies no financing structure for the alternative routes.

Table 11. Common-perimeter route comparison

t0 before investment; baseline full connection year 4, fixed end 17. All routes conditional and mutually exclusive.

Table 12. Early-access route assumptions

Early costs t1; early receipts years 2-3. Phase holding 50%; temporary holding 25%; temporary salvage t3. No renewable alternative-supply case.

Author calculations at the assumed 9% rate. Alternative supply is modelled only for demand assets; technical performance, consents and financing are unverified.

12 Locate committed exposure and date the financing obligation

The stage-exposure exhibit records an additional question: how much assumed nominal spending remains unrecovered if the sponsor exits after a commitment stage? Development recovery is zero. After equipment purchase, recovery is assumed to equal 20% of equipment spending; completion spending has no assumed recovery. Adding exit cost and deducting recovery gives unrecovered exposures after equipment of 52, 82 and 113. After full investment, the figures become 84, 132 and 193. These are undiscounted exposure measures under stipulated salvage assumptions. They are distinct from incremental option values and do not turn already spent amounts into new costs in a later investment decision.

The primary financing cases draw debt proportionately with the actual capital-spending stages. Maximum aggregate draws equal 60% of capital expenditure. Interest of 8.5% is capitalized on the existing balance at each construction year-end before that date's new draw. Ten annuity payments begin at the first operating receipt, with a target minimum CFADS-to-service ratio of 1.35. A separate coverage calculation can reduce the drawn proportion below the cap, although the 60% cap binds in all twelve stated financing cases. Actual lenders could impose other limits, draw conditions or interest treatment; these are conditional model terms.

For data-centre demand with two years of delay, the draws are 10.8 at t0, 67.2 at year 1 and 48 at year 3. Capitalized interest is 0.918 in year 1, 6.708030 in year 2 and 7.278213 in year 3. The total 14.904243 increases the service-start debt to 140.904243. Ten annual payments of 21.474892 then occur in years 4 through 13, giving operating coverage of 2.048904. Capitalized interest changes the financed claim without providing operating cash. The first operating year's interest is included in its annuity and is not capitalized again.

A one-payment reserve is sponsor-funded at the final construction date and earns no interest. Release after the last payment is assumed permitted. Data-centre sponsor funding over its two-year-delay construction period totals 113.874892: capital spending of 210 plus holding cash of 8.4 plus reserve of 21.474892, less draws of 126. This total comprises payments on several dates; it is neither a single t0 use nor an equity valuation. Its annual sponsor requirements are 7.2, 44.8, 4.2 and 57.674892. A commitment schedule would need to supply each amount when due, subject to separately evidenced financing permissions.

Table 13. Stage-dependent nominal exit exposure

Undiscounted sunk-cost exposure, not incremental NPV. Equipment recovery 20% assumed; no completion-cost recovery.

Author calculations in fictional USD millions. Undiscounted unrecovered exposure uses stipulated recoveries; already spent cash is excluded from a fresh forward choice.

Table 14. Draws and capitalized construction interest

8.5% interest; drawn proportionately with capex; service-start balance before first service-year accrual.

Table 15. Repayment follows actual energisation

Ten annual annuities; target 1.35; reserve equals one payment, funded before first operating year.

Table 16. Total dated funding without old debt

Sponsor plus draws equals capex plus holding plus reserve. Nominal multiyear totals, not t0 uses or equity NPV.

Table 17. Renewable two-year-delay amortisation

Payments years 4-13; no operating receipt assumed before year 4.

Table 18. Data-centre two-year-delay construction funding

Interest grows debt without a cash receipt; sponsor funds nondrawn costs, holding and reserve on the stated dates.

Author calculations in fictional USD millions. Repayment follows actual energisation in these conditional cases; sponsor totals combine separately dated payments.

13 Test the unchanged payment calendar and retain the model boundaries

A second financing diagnostic holds the original year-2 payment start fixed while operating receipts move later. It uses each no-delay annual service amount: 8.475088 for renewable export, 13.352744 for industrial demand and 19.343281 for data-centre demand. This is a conditional payment-obligation test, with one payment of reserve assumed available by year 1. It does not supply a second debt-draw and interest-restructuring model for the delayed investment. Its purpose is to expose what happens to the stipulated payment requirement when the operational and financing calendars separate. It must remain distinct from the primary grace-period financing cases.

With any positive delay, the fixed calendar includes at least one payment year with zero CFADS. Its minimum operating DSCR is therefore zero. If every service year must satisfy 1.35 and no alternative treatment is permitted, this structure supports zero positive debt. A reserve may pay an instalment while leaving the coverage calculation unchanged. With a three-year delay, the pre-receipt debt payments total 25.425263, 40.058232 and 58.029843. After using the one-payment reserve, further liquidity of 16.950175, 26.705488 and 38.686562 is required. Those amounts exclude construction costs, additional interest changes, fees and minimum operating cash. None is presented as committed support.

The operating valuation excludes debt draws, capitalized interest and reserve movements throughout. Financing exhibits reconcile these separately, preserving the distinction between economic value and a dated cash requirement. Closure payments remain in the operating NPV and outside the stipulated operating DSCR numerator. A satisfactory service-period ratio therefore gives no assurance that closure is funded; retained cash, a closure reserve or another separately identified source would require an additional schedule. Likewise, positive project NPV does not demonstrate lawful energisation, available supply, an executable relocation or a financeable route. Each conclusion is bounded by its stated assumptions.

The reproducible Python model produces eighteen exhibits with at most six columns, explicit annual cash-flow ledgers and five calculated figures. Decimal checks independently recompute 86 NPVs and 24 debt-balance or service values. Additional identities isolate nine fixed-end delay effects, nine retained-life effects, five early-access gains and nine holding-cost effects. Six break-even cash levels are substituted back into their valuation equations. Construction debt, amortization and cash sources-and-uses are separately reconciled. The charts show delay value, conditional access choices, stage exposure, financing burden and the unchanged-calendar coverage gap. They contain no empirical probabilities; transaction evidence must replace the fictional inputs before any actual investment, funding or operating commitment.

Table 19. Fixed-year 2 payment calendar diagnostic

One-payment reserve assumed. Further liquidity excludes construction costs and interest changes; delayed strict coverage debt capacity is zero.

Table 20. Retained scenario boundaries

Original fictional analytical definitions; no probabilities, commitments or observed outcomes.

Conditional diagnostic using no-delay service amounts and a reserve assumed available by year 1. This overlay does not recalculate delayed debt draws or interest restructuring. A reserve payment leaves operating DSCR unchanged.

14 Define the connection that the project can actually use

The proposed method begins by separating the connection required by a generating project from the connection required by a consuming facility. A generator seeks a route for exporting output subject to the conditions attached to that route. An industrial plant or data centre seeks an electricity supply that supports its proposed load and operating commitments. These are different commercial questions, even where the projects share a site or a network interface. The review should identify the direction of permitted power flow, the operating conditions assumed in the design and the contractual party responsible for obtaining the relevant connection. Any proposed combination of generation and demand should have its own interface assessment.

The connection agreement should be translated into a schedule of obligations and conditional milestones. The reviewer should identify the executed instrument, its amendments, the capacity to which it applies and the evidence required for each milestone. A stated connection date should be accompanied by its qualifications, including dependencies on third-party works and actions required from the applicant. The proposed review should distinguish a contractual commitment, a planning estimate and a date supplied informally during negotiations. An application or an offer under review should remain described according to its actual status. No general entitlement to capacity or priority is assumed in this method.

The meaning of firmness deserves specific attention. The project team should ask which interruptions, operating restrictions or network conditions the agreement permits and how those provisions affect the intended service. For an illustrative generator, a route with export restrictions may alter the timing and volume of receipts. For an illustrative data centre, restrictions on imported power may constrain usable computing capacity or require a different service commitment to customers. The legal and engineering reviews should identify the applicable limitations together. The financial model should then represent the operating consequence of those limitations without converting a connection rating into an assumed level of continuous commercial output.

The recommended first approval is a documented connection basis for the selected project configuration. It should state the permitted use, outstanding conditions and the cost and timing assumptions that management is authorised to develop. If the project changes its load profile, generating equipment or control arrangements, the team should reassess that basis before carrying forward the earlier conclusions. The decision record should retain the original proposal and the approved changes so that a later reviewer can determine which connection assumptions supported equipment orders or customer negotiations. This creates a practical boundary around the rights and capability included in the project's current investment case.

15 Reconcile the engineering and customer programmes at the interface

The next decision is whether the delivery programme supports the date on which the business proposes to begin operating. A recommended programme should connect network works, customer installations and the tests required before the intended power flow can begin. Each interface should have a named deliverable and an acceptance authority. The parties should establish who designs and installs the relevant equipment, who witnesses tests and what information must be supplied before energisation. These recommendations require agreement with the relevant counterparties; they do not assume a standard division of responsibility between every network operator and project customer.

The engineering review should identify dependencies that can change the order of work. A hypothetical project could complete its buildings before a network reinforcement is available, while its own protection settings still require coordination with the final network design. The programme should represent both dependencies and identify which activity determines the earliest achievable operating date. The team should also distinguish equipment delivery from installation, successful testing and accepted operation. A purchase order showing a dispatch date supports a procurement milestone. Further evidence is needed to support a claim that the corresponding system will be ready to supply the required service on that date.

Generation and demand projects should carry different commissioning questions into this programme. A generating project should examine the conditions under which it may export during testing and whether any resulting receipts are available under its commercial arrangements. A demand project should examine the power needed for commissioning, the transition to its operating load and any limits on simultaneous equipment starts or staged occupancy. In each case, the project should investigate how temporary operating conditions differ from the intended permanent arrangement. The model should retain any resulting expenditure or revenue limitation rather than assuming that first energisation represents full commercial availability.

The proposed interface review should conclude with a jointly reconciled list of unresolved dependencies and their effect on the operating date. Where an activity has no reliable completion estimate, the programme should identify the investigation needed to develop one and the decisions that remain exposed in the meantime. A contractor's revised schedule should be assessed against the customer's commissioning and service commitments before it is accepted as the financial planning case. Management can then approve expenditure against a programme that explains its dependencies and limitations. Subsequent monitoring should compare completed acceptance events with that programme, preserving a clear record of why the forecast operating date has changed.

16 Test phased operation and alternative supply as distinct designs

Phasing can be considered when the proposed initial operation uses less network capacity than the completed project. The review should establish which equipment and services can operate within that initial boundary and whether the relevant counterparties permit the proposed arrangement. A phased data centre case should connect available supply to the specific customer capacity that can be commissioned and supported. A phased generating project should identify the equipment capable of exporting, the relevant operating restrictions and any shared systems needed from the later phase. The financial comparison should show the expenditure required for the phase and the commitments deferred until the remaining connection conditions are resolved.

Behind-the-meter generation should receive a separate technical and contractual assessment. In an illustrative demand project, local generation could be proposed to support part of the site's requirements. The analysis should investigate the applicable operating permissions, fuel or energy supply, maintenance arrangements and the network interface. It should identify whether export is proposed, whether parallel operation is permitted and how the design responds when local generation is unavailable. The project should also distinguish the operating role assumed for local generation from the role required by the customer's service terms. None of these permissions or capabilities follows automatically from locating equipment on the customer's site.

Storage should be assessed against a defined operating duty. The proposed review should specify how it is charged, the intended discharge pattern and the limits relevant to the required service. The engineer should examine the stated ratings, usable energy and operating constraints under the selected design. The commercial team should identify the assumptions governing charging electricity, replacement expenditure and any network services revenue included in the case. A storage installation proposed for a short interruption requires a different assessment from one expected to cover an extended restriction on imported supply. The model should represent the selected duty explicitly and identify where operation depends on a supply source that is itself uncertain.

The recommended decision is a conditional design approval supported by revised operating and funding cases. The team should compare the phased or alternative arrangement with the original configuration using the same service boundary. Additional fuel costs, operating staff, maintenance, testing and eventual removal or integration should be included where they apply. Proposed income from spare capacity should remain conditional on verified rights and commercial terms. Before ordering equipment, the board should receive the required consents, unresolved design questions and the consequences of rejection or delay. This assessment gives management a basis for pursuing an alternative arrangement without treating a technically plausible configuration as an approved connection solution.

17 Fund connection obligations on the dates they become payable

The proposed funding schedule should follow the payment obligations in the relevant agreements and orders. Deposits, progress payments and security should be recorded with their due dates, conditions and funding sources. The model should identify amounts that are payable before the project can draw its proposed financing or receive operating revenue. A total capital-cost estimate leaves that timing question unresolved. Where the timing remains uncertain, the schedule should show a clearly identified planning assumption and the event that could change it. Management should be able to distinguish a future funding intention from funds that can be used to meet a specific obligation.

Deposits and security require separate treatment in that schedule. The review should establish whether a payment is refundable, the conditions for release and the circumstances in which it may be retained or drawn. Any proposed bank instrument should be assessed against its actual terms, including collateral requirements and expiry. A deposit shown as recoverable in an eventual project valuation can still create an earlier cash requirement. The decision paper should identify the party entitled to any repayment and the obligations that must be satisfied first. No universal refund right or release timing is assumed for connection-related payments in this method.

The financing case should then be reconciled with the connection and construction programmes. In an illustrative project, a lender might require evidence of a particular connection milestone before funding specified costs. The review should use the actual proposed financing terms to determine whether that condition creates a gap before a network payment falls due. The board should see the amount and duration of any sponsor funding proposed to cover the gap, together with the downside if the milestone is delayed. Financing fees and continuing commitments should be dated consistently with the scenario being tested. A revised operating date should trigger a review of these cash requirements.

The recommended commitment decision should include a defined termination or withdrawal case. The project should identify payments already made, remaining contractual obligations and the evidence supporting any recovery assumption. The analysis should include the cost of unwinding equipment orders and site commitments where relevant, subject to the executed terms. Management should also investigate whether a revised project scope changes the treatment of amounts previously paid. The board can then compare the additional expenditure required to continue with the cash consequences of stopping or reducing the project. This comparison should remain current at each material payment date, particularly when funding is being committed ahead of unresolved connection conditions.

18 Align customer promises with the available operating capability

Customer revenue assumptions should be built from the service the project can support under its proposed connection and operating design. For a generator, the review should reconcile export capability with the delivery obligations, payment provisions and operating restrictions in its commercial arrangements. For an industrial facility, the analysis should examine how electricity constraints affect production and the ability to meet customer orders. For a data centre, it should connect the proposed supply arrangement to the capacity and service characteristics offered to customers. These applications require project-specific evidence; a connection milestone alone does not establish that the full revenue case is available.

The proposed commercial review should identify commitments that become costly before the corresponding capability is ready. An illustrative customer contract might contain a service commencement date, a staged capacity commitment or a condition permitting the customer to withdraw. The project team should record those terms accurately and investigate the consequence of a delayed operating date. Where discussions remain preliminary, the model should identify the commercial assumptions and avoid treating expressions of interest as contracted receipts. The board should receive a reconciliation between the capacity included in customer commitments, the capacity supported by the engineering programme and the capacity assumed in the revenue model.

Alternative supply arrangements should be assessed against customer requirements before they are included in a continuity case. A customer may specify operating characteristics or other conditions that require review alongside the technical design. The commercial team should establish which proposed changes require notice, consent or amended terms under the actual agreement. If temporary operation supports a narrower service, the model should use the corresponding revenue and cost assumptions. Any proposed compensation, price adjustment or customer concession should be identified as a negotiation position until agreed. This treatment allows the board to understand the commercial cost of maintaining a service while connection conditions remain unresolved.

The recommended approval is a customer commitment boundary tied to the current evidence. Management should be authorised to offer only the capacity, commencement conditions and service terms supported by that boundary, with defined escalation for exceptions. A proposed new customer commitment should be tested against existing obligations so that the same constrained capacity is not allocated inconsistently across the commercial plan. Following a programme change, the project should review customer communications and the resulting cash implications before retaining the earlier revenue forecast. The board can then assess the value of accelerating readiness against the cost and risk of commitments that depend on an unconfirmed operating capability.

19 Stage board expenditure around observable progress

The proposed board process should identify the next decision that requires money and the evidence needed to support it. Early expenditure may be justified where it resolves a material design, connection or customer question. Later commitments should be assessed against the results of that work and the remaining dependencies. The paper should state what management is authorised to spend, the purpose of that expenditure and the circumstances requiring renewed approval. This approach can be applied to generation, industrial demand and data centre projects while preserving their different operating and commercial requirements. It assumes no standard sequence of statutory or contractual approvals.

A useful approval paper should explain the consequences of both proceeding and waiting. An illustrative equipment order may preserve a proposed installation date while exposing the owner to cancellation costs if the connection basis changes. The board should receive the order terms, the evidence supporting the required specification and the alternatives available if approval is deferred. Where the analysis relies on an estimated delivery period or unconfirmed consent date, the assumption should be identified alongside its effect on expenditure and customer commitments. The recommendation should make the proposed exposure explicit so that a later review can compare the decision with the evidence available at the time.

Monitoring should focus on accepted deliverables and changes affecting the decision. A proposed monthly report should distinguish completed work, work awaiting acceptance and revised forecasts. It should reconcile changes in connection conditions, engineering scope, payment dates and customer obligations with the approved case. A milestone reported as complete should identify the acceptance evidence and any remaining qualification. If a material condition is missed, the report should show which expenditure can still be deferred and which obligations continue. The purpose is to give the board a current choice about further commitment, supported by a dated view of the project's remaining funding and delivery requirements.

The final operating approval should reconcile the connection actually available, the tested installation and the service management proposes to provide. The record should identify outstanding restrictions and the person responsible for monitoring them after commencement. Where the project begins in phases, the approval should preserve the boundary between the accepted initial operation and later capacity requiring further evidence. The board should also retain a record of material assumptions that changed during development, including the resulting expenditure and customer consequences. This provides a practical basis for reviewing the investment decision and for approving subsequent expansion without carrying forward connection or revenue assumptions that no longer describe the operating project.

20 Convert the comparison into the next authorised commitment

The proposed board recommendation should select a next action with an amount, an evidence condition and a reconsideration date. It should name the alternative preserved by that action and the alternative foreclosed by it. A connection study may produce information useful to more than one design. An equipment order may commit to one configuration and carry cancellation exposure. Each proposed expenditure should state which result management expects to obtain and how that result changes the subsequent choice.

Sensitivity analysis can direct the next investigation. If a route becomes preferable only above a certain operating-cash level, obtain evidence for the customer volume and cost assumptions that determine that level. If its advantage depends on a shorter connection interval, request the network and equipment evidence supporting that interval. The numerical break-even is an author-created decision threshold. It supplies no proof that a technical route, consent or customer price is available. The committee should retain the unresolved prerequisite beside the numerical result.

At portfolio level, the proposed review should group projects by shared delivery dependencies. Several sites can rely on the same reinforcement, supplier production slot or sponsor facility. A common-delay scenario can be tested without assigning an empirical correlation that has not been measured. Aggregate the resulting capital calls by date and compare them with the funding actually available under the sponsor's constraints. A separately incorporated project remains exposed to a shared cash shortage when its funding source depends on the same constrained balance sheet.

The final recommendation should include a funded stop, pause or reduction case. State the obligations that continue if the next milestone fails, the evidence needed for any recovery and the latest date at which management can preserve an alternative. After a network, financing or customer change, reconcile the revised decision with the original approval and explain the cash difference. This proposed discipline gives the connection issue an explicit place in valuation, financing and execution. The paper's worked examples remain fictional; an actual commitment requires verified rights, technical capability, commercial terms and usable funds.

References

1. Federal Energy Regulatory Commission. Explainer on the Interconnection Final Rule. 2025-01-23. Introduction; The Interconnection Final Rule; First and Second Key Areas. https://www.ferc.gov/explainer-interconnection-final-rule
2. Federal Energy Regulatory Commission. Explainer on the Interconnection Final Rule Requests for Rehearing and Clarification (Order 2023-A). 2024-03-21. Acceptable Forms of Security; What did FERC decide?. https://www.ferc.gov/explainer-interconnection-final-rule-2023-A
3. Federal Energy Regulatory Commission. FERC Launches Aggressive Targeted Action to Speed Large Load Integration. 2026-06-18. Opening paragraph; Tailored Orders, Five Categories of Reform. https://ferc.gov/news-events/news/ferc-launches-aggressive-targeted-action-speed-large-load-integration
4. US Department of Energy. Transmission Interconnection Roadmap. 2024-04. Executive Summary printed pp. viii-ix; printed p. 2 footnote 10; printed p. 7 target definition (PDF pages 8-9,23,28). https://www.energy.gov/sites/default/files/2024-04/i2X%20Transmission%20Interconnection%20Roadmap_1.pdf
5. Lawrence Berkeley National Laboratory and GridTracker. Queued Up: 2026 Edition; official summary. 2026-06. Scope paragraphs and Key highlights from Queued Up 2026. https://emp.lbl.gov/queues
6. National Energy System Operator. Evidence handbook and other G2 WQ submission resources. n.d. What are Gate 1 and Gate 2?; Evidence based approach; page notice. https://www.neso.energy/industry-information/connections-reform/evidence-handbook-and-other-g2wq-submission-resources
7. National Energy System Operator and network delivery partners. Connections reform timeline. 2026-02-13. Timeline Update; offer timeframes; FAQ on technical matters; Ofgem confirmation. https://www.neso.energy/industry-information/connections-reform/connections-reform-timeline
8. Ofgem. Strategic energy planning and connections reform in 2026: putting plans into action. 2026-04-21. Connections queue reform; Wider connections reforms. https://www.ofgem.gov.uk/blog/strategic-energy-planning-and-connections-reform-2026-putting-plans-action
9. Ofgem. Proposed data centre connection reforms. 2026-07-29. Publication/status block; Consultation description; proposed milestones. https://www.ofgem.gov.uk/consultation/proposed-data-centre-connection-reforms
10. Ofgem. Enabling greater self-build of transmission connections. 2026-10-07. Publication/status block; Consultation description. https://www.ofgem.gov.uk/consultation/enabling-greater-self-build-transmission-connections
11. International Energy Agency. Energy and AI: AI and energy security. 2025-04. Connecting data centres to electricity grids. https://www.iea.org/reports/energy-and-ai/ai-and-energy-security
12. International Energy Agency. Data centre electricity use surged in 2025, even with tightening bottlenecks driving a scramble for solutions. 2026-04-16. Physical bottlenecks paragraph following electricity-demand discussion. https://www.iea.org/news/data-centre-electricity-use-surged-in-2025-even-with-tightening-bottlenecks-driving-a-scramble-for-solutions

Public sources accessed 9 October 2026. Model inputs, calculated examples and proposed decision records are author-created illustrations. No actual project mandate, connection entitlement, technical certification, financing commitment or market valuation is represented.

About the Author

Chennakeshav (CK) is a corporate finance and investment banking executive with 25+ years of global experience in deal origination, structuring and execution across M&A, growth capital and corporate strategy. He has led value-creation mandates for founders, corporates and funds — bridging the boardroom view to hands-on execution and close.

His career spans Morgan Stanley, HSBC, Lloyds Banking Group, EWEC, ADQ portfolio companies and Emirates Growth Fund, across TMT, real estate, fintech, deeptech, cleantech, infrastructure and energy. He has partnered with C-suite leaders, private equity and venture funds, sovereign wealth funds and family offices to finance complex fund raises and scale-up ventures, and has led M&A due diligence, post-merger integration and business-transformation initiatives to create value.

At Matchpoint Partners he is Managing Partner, leading the firm's corporate finance, M&A and capital-raising practice. He holds an MBA from London Business School, an engineering degree from VTU and a Master of Laws (LLM, in progress) from UCL London.

An active start-up mentor, CK mentors at Techstars, DIFC FinTech Hive, Startup Grind, Founder Institute and IN5, serves as Entrepreneur Mentor in Residence (EMiR) at London Business School, and judges the Entrepreneurship World Cup.

https://www.linkedin.com/in/ckadya/ https://www.matchpoint-partners.com/team/ck-adya.html