Data Centre Model Validation
Executive Summary
Key Takeaways
- ✓ Data centre model validation tests whether input assumptions are reasonable and well-sourced, distinct from a model audit, which tests whether the model's formulas are structurally correct.
- ✓ Capacity headroom assumptions should be validated against independent engineering or facility survey evidence, not accepted from the model owner's own reported figure without corroboration.
- ✓ PUE assumptions should be validated against actual metered facility data or credible engineering projections specific to the facility's climate and cooling technology, not a generic industry-average figure.
- ✓ Sensitivity coverage should be validated as adequate if it tests the highest-impact drivers identified in sensitivity analysis, occupancy, price, power cost, and PUE, not merely a token range applied to a few arbitrarily chosen variables.
Objective¶
This guide sets out how to validate the input assumptions of a data centre financial model within Data Centre Financial Modelling, distinct from the structural formula testing covered in Data Centre Model Audit.
Validation Versus Audit¶
Model validation tests whether the model's input assumptions are reasonable and well-sourced. A model audit tests whether the model's formulas are structurally correct. Both are necessary, and a model can pass one while failing the other, structurally correct formulas can still be fed unreasonable assumptions, and reasonable assumptions can still be miscalculated by a structurally flawed formula.
Validating Capacity Headroom Assumptions¶
Capacity headroom assumptions should be validated against independent engineering or facility survey evidence confirming actual remaining power, space, and cooling headroom, consistent with Data Centre Capacity Planning Models, rather than accepted from the model owner's own reported figure without independent corroboration.
Validating PUE Assumptions¶
PUE assumptions should be validated against actual metered facility data or credible engineering projections specific to the facility's climate and cooling technology, not a generic industry-average figure, since PUE varies materially by these facility-specific factors and a generic assumption can materially misstate power cost.
Validating Pricing and Contract Term Assumptions¶
Pricing assumptions should be validated against the facility's actual achieved-rate track record and, secondarily, competitive benchmarks, and contract term assumptions, escalators, renewal rates, SLA provisions, should be validated directly against the underlying contracts rather than a summarised or assumed representation of their terms.
Validating Sensitivity Coverage¶
Sensitivity coverage should be validated as adequate if it tests the highest-impact drivers identified through Data Centre Sensitivity Analysis, occupancy, price, power cost, and PUE, with a realistic range for each, rather than a token sensitivity range applied to a small, arbitrarily chosen set of variables that may not include the drivers actually most material to the outcome.
Common Validation Findings¶
Recurring findings include: capacity headroom figures accepted without independent engineering corroboration; PUE assumptions sourced from a generic industry average rather than the facility's actual metered data or climate-specific engineering projection; and sensitivity analysis covering only a narrow, low-impact set of variables while omitting the drivers sensitivity ranking would identify as most material.
Continue Reading¶
Related Pillars¶
Related Technical Guides¶
Related Glossary¶
How OXXON tests thisRun a free structural check with FMAE
Frequently Asked Questions
What is the difference between data centre model validation and a model audit?
Model validation tests whether the model's input assumptions, capacity, pricing, PUE, contract terms, are reasonable and well-sourced. A model audit tests whether the model's formulas are structurally correct, computing what they are represented to compute. The two are complementary but test different things.
How should capacity headroom assumptions be validated?
Against independent engineering or facility survey evidence confirming actual remaining power, space, and cooling headroom, not accepted from the model owner's own reported figure without independent corroboration, since the reported figure may not reflect the facility's actual binding constraint.
How should PUE assumptions be validated?
Against actual metered facility data or credible engineering projections specific to the facility's climate and cooling technology, not a generic industry-average figure, since PUE varies materially by these facility-specific factors and a generic assumption can materially misstate power cost.
What makes sensitivity coverage adequate in a validation review?
Testing the highest-impact drivers identified through sensitivity analysis, occupancy, price, power cost, and PUE, with a realistic range for each, rather than a token sensitivity range applied to a small, arbitrarily chosen set of variables that may not include the drivers actually most material to the outcome.
Related Articles
Data Centre Financial Modelling
Data centre financial modelling is the discipline of modelling a data centre operator's revenue, cost, and capital structure from its capacity-denominated drivers, power, space, and cooling capacity, rack density, and tenant contract structure, rather than the generic market-price and headcount-growth drivers used in most corporate models, or the pure occupancy-and-lease-term drivers of conventional commercial real estate. This page is the hub for the Knowledge Centre's data centre financial modelling content: how colocation, hyperscale, and enterprise business models each require a distinct model architecture, how rack revenue and occupancy are decomposed into their separable underlying drivers, and how capacity planning and financial KPIs tie the model together, as this domain expands to cover operations, revenue, investment, and governance practice across the sector.
Data Centre Model Audit
A data centre model audit tests the structural integrity of the model's formulas and logic, distinct from validating the reasonableness of its input assumptions. This guide sets out the audit procedures specific to a data centre model: verifying capacity constraint calculations, revenue driver formulas, and power cost pass-through logic actually compute what they are represented to compute, free of circularity, hardcoding, or broken links.
Power Usage Effectiveness (PUE)
Power usage effectiveness (PUE) is calculated as total facility power divided by critical IT load power, with a value approaching 1.0 indicating that nearly all power consumed is delivered to IT equipment rather than lost to cooling, power distribution, and other non-IT overhead. PUE is the standard industry measure of data centre power efficiency, and because power is typically one of the largest operating cost categories, a facility's PUE directly drives its power cost per unit of billable capacity and, in turn, its profitability.
Data Centre Sensitivity Analysis
Data centre sensitivity analysis flexes one driver at a time, holding all others constant, to rank which individual assumptions, occupancy, pricing, power cost, and PUE, most affect model outputs such as revenue, EBITDA, or debt service coverage. This guide sets out how to construct a driver-by-driver sensitivity table for a data centre model and how it complements, rather than substitutes for, correlated scenario analysis.