Prompts that build the model
Most people ask their AI for a financial model and get whatever it improvises. These prompts hand it the structure I’ve refined over twenty years of building these models, so what comes back has the right inputs, formulas that trace back to them, and checks that tell you when something’s off.
They’re free. Paste one into Claude, ChatGPT or Gemini, answer its questions, and it builds the workbook in my house style. Rather start from something finished? Every model is open source and pay what you want.
For founders
Forecast the business, check the unit economics, and plan the round.
LTV, LTV to CAC and payback for one customer, across recurring, one-time and transaction revenue.
Subscribers, MRR and ARR, billings and deferred revenue, with contract and billing cycles.
Monthly customer cohorts, repeat orders from a retention curve, contribution margin and LTV.
Driver-based revenue and costs, three linked statements, cash and months of runway.
A full operating model, built in four checked stages: revenue engine, hiring plan, statements, metrics.
A priced round with SAFEs, notes and the option pool, then who gets what at each exit value.
For investors
Value a single deal, or model a whole fund from its terms.
A priced round with SAFEs, notes and the option pool, then who gets what at each exit value.
One investment from seed to exit, with dilution, graduation odds, a waterfall and expected value.
Fund size, fees and portfolio construction, run by year to LP and GP returns, with three scenarios.
Built one with AI?
A model that looks right isn’t always right. Bring it to a working session and we’ll go through it together.
Book a working sessionNeed a different model?
Tell me what you’re building. The next prompts come from what people ask for.
Tell me what you need