Back into the max land price without inventing a number
Rung 4 of 10 · Constraints + negative instructions
1 · Learn the move · Constraints + negative instructions
Residual land value is the discipline that keeps you from overpaying for dirt: gross sellout minus every cost, financing, selling, contingency, and your required profit leaves the residual, the most you can pay for the land. The temptation is to let the AI 'plug in reasonable numbers', and that is how a bid becomes a fantasy. The move is constraints plus hard negatives: the AI builds the STRUCTURE, each line a [YOUR NUMBER] slot fed from your own budget and market study. The negatives are absolute, never invent a finished value, a hard or soft cost, an absorption pace, or a profit margin. The residual itself is a subtraction chain you do not run in the chat; it belongs in your model, tagged rather than computed. Label every assumption. The max land price is your model's output, not the AI's, and none of this is investment advice.
What's the most I should pay for this land? It yields about 80 lots or pads and I want a good margin.
2 · Your turn. You write the prompt
You're a land-acquisition manager backing into the max price for a raw parcel that yields roughly 80 finished lots or pads. You'll supply gross sellout, hard-cost and soft-cost figures, financing, selling costs, contingency, and your required developer profit from your own budget and market study. You want the residual FRAMEWORK built now, before every figure is in, so you can drop numbers in, WITHOUT the AI inventing any value, cost, absorption pace, or margin, and WITHOUT it computing the residual itself. Write that prompt.
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.