Bottom line: A data center underwriting platform should model power—$/kW/mo × MW × ramp—with live sensitivity, phased capex, and LP-ready outputs. If it still prices on $/SF, it may misstate your risk.
Underwriting: power, not square feet
Hyperscale rent is sold by the kilowatt. A typical 32MW deal swings $1.92M/year for every $5/kW/mo move. Generic CRE tools miss this because they model PSF.
An investment-grade platform models the power schedule—contracted, ramped, then billed—and ties opex, occupancy, and tenant mix to it.
Six-point buyer checklist
- Power-centric rent roll: Leases on $/kW/mo, with TCV, escalations, and ramp built in.
- Live sensitivity: Cap rate, lease rate, and timing sliders update IRR/EM instantly for IC.
- Phased capex linkage: Shell, MEP, and expansion spend feeds invested capital and returns, rather than a side sheet.
- Audit trail and versioning: See who changed what and when, with one-click undo.
- Portfolio view: Underwrite one deal and see the fund impact.
- Investor output: Generate the waterfall and LP report from the same model, rather than rebuilding them in Excel.
A claim attributed to Univ. of Hawaii says 88% of spreadsheets contain errors; in a data center waterfall, an error can affect distributions, not just rounding.
Why DIJASOFT versus generic tools
DIJASOFT is data center investment management software purpose-built for digital infrastructure investments. It models CBRE/JLL 2024 benchmarks—hyperscale at $130–150/kW/mo versus retail colo at $250+—natively, and hyperscale, GPU, and retail colo can sit in one portfolio with opex that responds by tenant type.
Next: See the full evaluation checklist and comparison table on the buying-keyword pillar page: Data Center Investment Management Software →.
*Cluster: Pro Forma Software: Excel vs Platform · Hyperscale: The $/kW Mistake*