Eleven months of runway, not eighteen
He thought he had eighteen months. He had eleven. The gap was every assumption he never wrote down.
In January a founder told me his runway ran a year and a half. By July it was four months. He hadn't lost a customer, missed a launch, or booked a bad month. The year and a half was never real. It was a number computed correctly against assumptions he had never written down.
Our model assumed new ARR would land on schedule; it came in at 60%. It assumed the next two hires would land on plan; they landed three weeks late at higher comp. It assumed the software stack would hold flat; it grew month over month. It assumed no surprise wires; two cleared. Each miss was minor alone. Stacked, they ate more than half a year of cash without once showing up in a board update.
Runway is the number founders trust most and audit least. It feels precise because it has a denominator. In practice it's the average of a stack of optimistic assumptions, printed as one integer because an integer is easier to say out loud. The integer is also the part that turns out wrong.
A truer answer isn't a number. It's a range across three scenarios. One, the plan you built. Two, the plan minus your top customer. Three, the plan with your next milestone slipping two months. The figure to lead with, to the board, to yourself, to the person you're about to hire, is the worst of the three. Anything else is a wish counted in months.
Burn lies in its own direction too. Headcount is the lagging line everyone watches. The leading signals are quieter: vendor renewals creeping up, the stack adding a subscription here and there, one-time costs nobody flags as recurring because technically they aren't. By the time the headcount line moves, the burn rate has been wrong for two quarters. The same drift shows up in forecasts that degrade as the company grows.
The founder eventually raised a bridge at terms he hated, because he had no choice. At the runway he believed in, the bridge was unnecessary. At the runway he had, it was the only move. The gap wasn't bad luck. It was the predictable cost of trusting an integer, the same way a forecast gets mistaken for a promise.
The cheapest finance upgrade most companies can make is to stop reporting runway as a number and start reporting it as a range. The story you tell the board gets less confident. The story you tell yourself gets accurate. Accurate is what survives the quarter.