Verified against ChatGPT · 2026-08-09
Project cash runway from a bank balance and known inflows/outflows before a hard funding or hiring decision
Builds a week-by-week or month-by-month cash flow projection from a starting balance and known future transactions, surfacing the exact date the balance would go negative under stated assumptions so a runway decision isn't based on a gut feel.
The prompt
Ready to copy — highlighted parts are example details you can swap.
You are building a cash flow projection to answer one question precisely: given the current balance and known future inflows and outflows, when does the balance run out, or how much buffer is there against that? This is a structuring tool to organize the requester's own numbers, not a substitute for advice from an accountant or financial advisor on what decision to make with the result. STARTING CASH POSITION $84,000 in the operating account as of August 1. KNOWN INFLOWS Client invoice #1042, $22,000, due and expected August 15. KNOWN OUTFLOWS Payroll $38,000 on the 1st and 15th each month, office rent $4,200 on the 1st, software subscriptions $1,900/month. PROJECTION HORIZON 16 weeks, through late November. UNCERTAIN ITEMS A second client contract worth $15,000/month that may or may not close by September — call it 50/50. BUILD RULES Project the running balance period by period (weekly if the horizon is under 3 months, monthly otherwise) and show the balance after each period, not just a final number — the point of a runway projection is seeing the trajectory, not just the endpoint. For every uncertain item (a deal that might close, a client payment that might be late), build the projection using the conservative case as the primary scenario, and separately note what the balance would look like if the optimistic case held instead — never blend an uncertain inflow into the main projection as if it were confirmed. If the balance is projected to go negative or below a reasonable minimum buffer at any point in the horizon, state the exact period this happens in, not just "runway is tight." Do not recommend a specific action (cut costs, raise a round, delay hiring) — surface the numbers and the date they imply, and let the requester or their advisor decide what to do about it. OUTPUT FORMAT 1. Period-by-period table: period | inflows | outflows | net | running balance. 2. Conservative-case runway date (if applicable) stated as a specific period. 3. Optimistic-case comparison, if uncertain items were provided. 4. A one-line summary of which single assumption most changes the outcome if it's wrong.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
Requiring the model to show the running balance after every period rather than just a final runway number matters because a single end-of-horizon figure hides the actual shape of the risk — a business can look fine on a 16-week average while dipping dangerously low in week 6 because payroll and rent land the same week an expected invoice slips, and that dip is invisible unless the trajectory is shown period by period. The conservative-versus-optimistic split for uncertain items exists specifically because GPT-5.1, like most language models, will otherwise average an uncertain 50/50 inflow into the main projection as a half-weighted number, which produces a balance figure that doesn't correspond to either real-world outcome — a deal either closes or it doesn't, and blending it produces a phantom middle scenario nobody should plan around. Refusing to recommend a specific action (cut costs, raise money, delay a hire) keeps the tool inside its actual competence: the model can compute what the numbers imply given the inputs, but which trade-off to make in response depends on context the model can't see — the founder's risk tolerance, what a lender or investor would think, what's contractually committed — so recommending a decision would be presenting a guess as advice. Naming the single most decision-relevant assumption last gives the requester one thing to sanity-check with their accountant instead of a wall of numbers with no sense of which one actually matters most.
What you get back
Week of Aug 15: inflow $22,000, outflow $38,000 (payroll), running balance $68,000. ... Week of Oct 3 (conservative case, second contract excluded): running balance dips to $6,200 before the next invoice lands. Optimistic case (second contract closes in Sept): balance stays above $40,000 throughout. Most decision-relevant assumption: whether the second contract closes before October payroll.
Verified against
ChatGPT GPT-5.1 · 2026-08-09
Changelog
- 2026-08-09 — Initial publish, verified against ChatGPT GPT-5.1.
Need this built into your business?
If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.
EXPLORE WHAT SCULT BUILDS
