Asking vs effective. Survey the rent that's actually paid
Rung 2 of 10 · Comparison matrix
1 · Learn the move · Comparison matrix
Asking rent is the sticker; effective rent is what the landlord nets after free rent and TI. This is a MARKET survey, you read what a whole submarket actually rents for, so the deliverable LEADS with the market read, not the pricing of one space (that's the leasing broker's job). A comparison matrix lays each comp side by side, asking, term, concessions, effective, so the gap between headline and reality across the submarket is visible. The trap is letting AI fabricate an effective-rent number or fill a concession it never saw. So make it place only pasted figures, show the effective-rent arithmetic as a setup tagged [verify] rather than a silently computed total, and mark any missing input UNKNOWN. A comp missing its TI can't yield an effective rent. That cell stays UNKNOWN, not a plausible fill. You get a survey you can defend line by line.
Turn these lease comps into an effective rent comparison and fill in anything I'm missing.
2 · Your turn. You write the prompt
You're building a rent survey for an office submarket. You paste three VERIFIED lease comps plus the subject's asking: Comp A asking $38.00/SF, 7-yr term, 6 months free, TI $65/SF; Comp B asking $41.50/SF, 10-yr term, 10 months free, TI $85/SF; Comp C asking $36.00/SF, 5-yr term, 4 months free, TI not provided; Subject asking $39.00/SF. Write a prompt that lays these into an asking-vs-effective matrix, shows the effective-rent arithmetic as a [verify] setup on the complete comps, marks Comp C's missing TI (and its effective rent) UNKNOWN, and invents no comp or concession.
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.