Runway is months of cash left, calculated as cash on hand divided by average monthly net burn — and the honest version uses net burn (cash out minus cash in), gross margin timing, and unpaid obligations, not the flattering balance your dashboard defaults to. The median seed-stage startup that fails does so around month 20 to 24 of its life, with running out of cash the top stated cause, per a 2024 analysis of failed-startup post-mortems published by CB Insights-style industry research. This is a guide to the arithmetic, not financial advice.
Most founders know the formula and still get the number wrong, in ways that run in one direction: optimistic. Here are the four places the error hides.
What is the difference between gross and net burn?
Gross burn is total cash out per month. Net burn is cash out minus cash collected in. Net is the number your runway actually runs on, because revenue that arrives reduces how much cash you spend. A company spending $120,000 a month and collecting $70,000 has a gross burn of $120,000 and a net burn of $50,000 — the same bank balance supports less than half the runway at gross burn that it does at net. Use net burn for runway, and know both numbers: gross burn is what a cost-cutting lever moves, net burn is what growth spends.
Which cash balance should you divide?
The usable balance, which is lower than the ledger balance. Start from your bank total, then subtract what is already spoken for: unpaid contractor invoices you will honor, committed but unbilled expenses, payroll that hits on the first of the month, and any restricted cash such as deposits or held customer funds you cannot spend. The SBA's own cash-flow guidance for small businesses makes the same point in plainer terms: a cash-flow projection is only as good as its timing assumptions, per the agency's published templates.
One more subtraction, the one founders skip: the minimum operating buffer. Payroll and rent must clear even in your worst collection month, so treat roughly one month of fixed costs as untouchable and run the division on what remains.
How do you average the burn honestly?
Use a trailing three-month average, weighted toward the most recent month if your spending is growing. A single month's burn is noise — annual insurance payments, conference season, a recruiting batch — and the trailing average smooths it. If the most recent month's burn is 15 percent above the three-month average, use the recent month, because the trend is the truth and the average is the flattery.
Then stress the number. Recompute runway with revenue at 80 percent of plan, per the common practice of scenario planning recommended in published CFO guides for startups. If the answer falls under nine months, the finding is not academic: median time to close a seed round has run three to six months in recent market data, per Carta's published fundraising reports, so a raise started at six months of runway ends in a negotiation with no alternatives.
What does a runway dashboard actually look like?
| Line | Example figure | Notes |
|---|---|---|
| Cash in bank | $1,450,000 | Ledger balance, all accounts |
| Less: unpaid commitments | –$60,000 | Contractor invoices, upcoming payroll |
| Less: one-month fixed buffer | –$95,000 | Untouchable floor |
| Usable cash | $1,295,000 | Divide this |
| Trailing 3-mo avg net burn | $52,000 | Weighted to recent month |
| Runway | ~24.9 months | Recheck monthly |
The table's point is the two subtraction lines. A dashboard that divides the ledger balance by the best month's burn is not a dashboard; it is a mood.
How often should you recheck?
Monthly, on the same day, with the same lines — consistency matters more than precision. Recompute immediately after any event that changes the arithmetic: a raise, a hire batch, a price change, an annual contract won or lost. And when the number crosses 12 months, that is the planning trigger for what the next raise or the next spend program buys; when it crosses 9, it is the trigger to act.
What the arithmetic establishes: runway is a division, but each input has a honest version and a flattering one. What it cannot tell you is what your revenue does next — that is a forecast, and forecasts are where the discipline moves from math to judgement.