Hub 06 Cost and Value

A complete first-cost model for 1 MW of data center capacity, built from the ground up: long-lead equipment, construction, and soft costs, using benchmark figures ($13.6M per IT MW greenfield; LLE ~32% of project cost). The concrete worked example that grounds every abstraction that follows — when 06.20 talks lifetime cost and 06.50 talks business cases, this is the model they extend. Builds on: Hub 05 (the costed packages), Hubs 01-04 (what the packages are).

The full TCO model — CapEx plus lifetime OpEx — applied to power and cooling assets, using the Uptime Institute's true-TCO framework: IT load, cooling, and auxiliary energy broken out separately, because purchase price is typically only 20-30% of what an asset costs to own. Comparing a cheaper, less efficient unit against a premium one on ten-year economics instead of purchase price. Builds on: 06.10 (the walkthrough's first-cost model, extended to lifetime), 05.20 (should-cost), 01.20/02.10 (the assets).

How single points of efficiency become millions: a 1% UPS efficiency gain at 10 MW saves roughly $1.4M over ten years, and architecture choices compound through CapEx, maintenance, footprint, and electricity. Turning efficiency into procurement criteria — writing specs and evaluation weights that reward lifetime energy performance, not just first cost. Builds on: 06.20 (the TCO framework), 01.20 (UPS architectures), Hub 02 (PUE).

Applies 04.20's pricing structures to buying decisions: separating the reserved capacity fee from the metered energy pass-through, modeling what ramp does to total contract value, and reading wholesale vs. retail quotes the way a procurement expert does. This is 04.20 from the buyer's side of the table. Builds on: 04.20 — the pricing standard, now used as a buying tool.

Building the ROI case for bringing procurement capability in-house: cost of the internal function (people, systems, transition, knowledge transfer) against what it captures or protects (provider margin, direct-OEM pricing leverage, schedule certainty on long-lead equipment). Framed as a phased investment with decision gates, not a single bet — and quantifying the cost of delay, which in this market dwarfs the function's operating cost. Builds on: 06.20 (TCO), 06.40 (pricing), Hub 05 (what the function does).

The general method for justifying procurement investment to a CFO: category-by-category savings plans with realistic timing (roughly six months per strategic sourcing cycle), the investment required in resources, systems, and support, and pledged savings set against that investment. Picking quick-win categories with strong budget-holder buy-in first — and why underestimating execution time is the most common way these business cases fail. Builds on: 06.50 — generalizes the specific case.

Key sources: Uptime Institute on true TCO; GE Critical Power on energy efficiency; Data Center Knowledge (colocation battleground 2026); CEO Today on procurement transformation business cases.