APLD · IREN · WULF · CIFR · NUAI
Construction and Financing Risk in Data-Center Valuation
Why the path from powered land to stabilized NOI can matter more than the exit multiple.
Thesis
A practical framework for modeling capital draws, construction-period interest, contingencies, completion support, and dilution before stabilization.
Key chart
| Stress | Direct effect | Secondary effect |
|---|---|---|
| 10% capex overrun | More capital required | Higher debt or dilution |
| 12-month delay | Later cash flow | Interest carry and customer remedies |
| Higher loan spread | Lower project IRR | Lower debt capacity |
| Lower terminal multiple | Lower residual value | Tighter equity return |
The argument
The correct valuation unit is the full cash-flow path from today to stabilization, not stabilized NOI divided by a cap rate.
FactConstruction-period interest can accrue before a project begins producing revenue when debt is drawn during the build. InferenceApparently attractive stabilized economics can still produce a weak equity return when delay, cost overrun, or parent funding is material.Key findings
- Construction-period interest begins when capital is drawn, not when revenue starts.
- Unused capacity can carry land, staff, commitment fees, and common infrastructure costs.
- Project debt may be non-recourse in name but require parent completion support.
- Contingency and change orders should be modeled explicitly.
- DSCR, loan-to-cost, and minimum equity can bind simultaneously.
Counterarguments
A highly creditworthy tenant can fund or guarantee much of the build, reducing equity exposure. Standardized designs and shared infrastructure can also lower later-phase cost.
Risks
Equipment lead times, labor, interconnection, cooling, permitting, commissioning, customer acceptance, and refinancing can interact rather than occur independently.
Catalysts
Fixed-price contracts, equipment reservations, lender commitments, tenant deposits, and achieved construction milestones reduce specific risks.
Scenario analysis
Valuation analysis
Model monthly or quarterly capital draws, interest, fees, contingency, tenant contributions, and revenue ramp. Reconcile the financing sources and uses in every period.
Assumptions
Important fields include EPC structure, contingency, interest during construction, debt commitment timing, customer deposits, parent guarantees, DSCR, amortization, and refinancing.
Methodology
The project uses a sources-and-uses schedule before applying any stabilized cap rate.
Disconfirming evidence
The risk case weakens when the project is substantially pre-funded, fixed-price, standardized, and supported by strong customer obligations.
What would change the conclusion
A complete financing package and construction budget can move a project from speculative option value to underwritten phase value.
Primary sources
Use debt agreements, construction contracts where public, lease disclosures, permit records, and audited cash-flow statements. See the source archive.
Audit this conclusion
The conclusion can be summarized elsewhere. The full Ephesus Research page remains the place to inspect the calculation, evidence, sensitivities, revisions, and contrary evidence behind it.
Change the valuation assumptions
Adjust the discount rate, stabilized multiple, utilization, unit economics, funding mix, share count, delays, and completion probabilities.
Open exact sectionInspect every material assumption
Review evidence type, source label, date, confidence rating, and the note attached to each model input.
Open exact sectionCompare execution paths
Move between bear, base, and bull conditions, then inspect the phase-by-phase buildout schedule.
Open exact sectionStress-test the valuation
Open the complete sensitivity matrices for discount rates, terminal values, unit economics, delays, dilution, and execution risk.
Open exact sectionReview what changed
Open the dated revision record rather than relying on an undated excerpt or an older model output.
Open exact sectionDownload the underlying model
Open the public spreadsheet or machine-readable JSON and CSV representations for independent review.
Open exact sectionTest the conclusion against contrary evidence
Read the facts, limitations, and developments that would weaken, invalidate, or materially change the stated conclusion.
Open exact sectionTrace the evidence to its sources
Follow the source map to filings, company disclosures, contracts, permits, and other cited records.
Open exact sectionEvidence guide
Evidence and judgment labels
Statements marked Fact are intended to be directly supported by cited evidence. Guidance, estimates, assumptions, inferences, and speculation remain separately named so they are not mistaken for verified facts.
5
mapped sources
Yes
primary support
Related spreadsheets
Audit the linked model
IREN · AI infrastructure and digital assets
IREN Five-Year DCF and Buildout Model
This model estimates the cash IREN could generate from operating and planned sites, subtracts construction and hardware-replacement costs, reduces the value of uncertain projects, subtracts net debt, and divides what remains among diluted shares.
Question this model answers
What could one IREN share be worth at different stages of its AI and Bitcoin-mining buildout?
Base estimate per share
US$38.96
Outcomes shown
3
NUAI · Data centers and behind-the-meter power
NUAI TCDC Site Economics and Tenant Scenario Model
This model estimates the income each TCDC phase could produce, converts stabilized income into a project value, adjusts for NUAI's ownership and the chance that each phase is completed, subtracts net debt, and divides the result among diluted shares.
Question this model answers
What could NUAI's ownership in TCDC be worth if one or more phases are financed, built, and leased?
Base estimate per share
US$5.62
Outcomes shown
3
Evidence map
Mapped public sources
NUAI: A New Era of Compute
NUAI · Jul 5, 2026 · Texas
Relevant finding
Synthesizes TCDC's phased 1.4 GW plan, partner ecosystem, Phase 1 power question, behind-the-meter expansion, Macquarie timing, land contribution, GP/LP waterfall economics, construction financing, and execution risks.
Review notes
A 143-page secondary deep dive reviewed from the project's private NUAI research archive. It synthesizes filings, company materials, calls, counterparties, project economics, and author estimates. Independently verify every material claim and scenario input; do not treat the report as a substitute for primary evidence.
Relevant pages: Pages 13-14 for company and site overview; 19-36 for TCDC, partners and power; 46-52 for incentives and lease timing; 56-126 for JV and waterfall economics; 136-138 for the pipeline and New Mexico optionality.
Applied Digital regulatory filing — sample source entry
APLD · Jun 30, 2026 · United States
Relevant finding
Illustrative source for construction commitments, revenue recognition, debt, and share count.
Review notes
Replace with exact filing and debt footnotes.
Applied Digital investor presentation — sample source entry
APLD · Jul 1, 2026 · United States
Relevant finding
Illustrative source for contracted capacity and construction schedule.
Review notes
Separate signed commitments from prospective pipeline.
Version control
Article change log
Construction and Financing Risk in Data-Center Valuation
Jul 22, 2026 · APLD
Added construction-period interest and unused-capital carry.
Previous
Financing cost began at revenue start
Revised
Financing cost begins as capital is drawn
Reason
Construction funding creates negative carry before the asset produces NOI.
Source
Sample filing placeholder
Estimated effect
Reduced project IRR and increased required stabilized NOI.
Research status
Research status
Current
Conclusion
Mixed
Version
1.0.0-sample
Last reviewed
Aug 2, 2026
Access
Public and free
Continue reading
Related public research
IREN · NUAI · WULF
The Hidden Credit Link in the AI Power Boom
This report examines the credit channel linking frontier-model economics to data-center and power-development finance. Version 1.1 adds two explicit circuit breakers: proprietary providers can retain disproportionate monetization after losing token share, and stronger technology balance sheets can substitute for weaker standalone lab credit. The likely near-term result is financing bifurcation rather than uniform contraction.
Ucore Rare Metals
Ucore Five-Year Corporate SOTP: Louisiana Is the Core, but It Is Not the Whole Portfolio
The Louisiana SMC is Ucore's principal operating asset, but it is not the entire portfolio. This five-year SOTP separately values Louisiana, a risked Canadian Sm/Gd SMC, Bokan-Dotson Ridge, external RapidSX optionality, corporate cash and overhead, then models the funding and dilution required to reach commercial scale.
IREN
IREN: Five-Year DCF and Buildout Valuation
A current preliminary cohort-cash DCF mapped to the live Google Drive workbook. The analysis distinguishes IREN's official capacity stages, cohort financing, five-year primary GPU life, non-core post-contract tail economics, residual support, revenue sharing, and public community-model reference outputs.
Challenge a fact, formula, or interpretation.
Submissions are reviewed before publication. Credited contributors are listed only with permission; private email addresses are never displayed publicly.