Screen a data center on the one spec that decides it: power
Rung 2 of 10 · Comparison matrix
1 · Learn the move · Comparison matrix
Data-center flyers sell square footage and a big 'available MW' headline, but in 2026 land is easy and power is the gate, roughly $1.2T of deliveries chasing substation capacity that utilities commit years out. A comparison matrix with three honest values (FIT, NO-FIT, UNKNOWN) stops AI from reading '30 MW available' and treating it as delivered power. Uncommitted MW, no utility letter. Is a NO-FIT blocker, not a maybe. Cooling approach and PUE the flyer never states stay UNKNOWN. The UNKNOWN cells become your call sheet, and the real power-availability confirmation routes to the utility and an MEP engineer, not to you.
Is this data center a good fit for a hyperscale tenant? [PASTE]
2 · Your turn. You write the prompt
An investor hands you a flyer on a 500,000 SF building marketed at '30 MW available,' fiber from 2 carriers, and asks whether it clears a hyperscale requirement. The flyer names no utility commitment letter, no cooling approach, no PUE, no redundancy tier. You need a spec-by-spec screen before anyone books a site visit.
Remember: the AI sees only your prompt, not this page. If the situation isn't in your prompt, it doesn't exist.
Optional. These shape the output when you run your prompt below, not your score.