Flat runway arithmetic is the most comfortable wrong number in startup finance. Scroll and watch what the growth term does to a balance that felt safe on Monday.
Scroll and the months run forward. The line is the cash left in the account at the end of each month, and it starts at one point eight million with nothing spent yet.
Insurance limits, sweep terms and card underwriting rules all move, and the version in your head is usually the version from the last time you looked. One line per change, with the figure before and after.
Type your own figures. No sliders with a floor on them, because a company with 40,000 dollars in the bank has the same question as one with four million and deserves the same answer.
Spend growth is the term the flat formula drops, and four per cent a month is not a guess. It is what a normal quarter does to a 120,000 dollar cost base. Here it is taken apart.
So four per cent is on the low side, not the high side. And you do not have to trust our decomposition at all, because your own books already answer it. Put in what you spent twelve months ago and what you spend now, and the box below works out your actual rate.
The arithmetic is the twelfth root of now divided by then, minus one. Nothing clever. If that number is higher than the one you typed into the model above, the model above is optimistic and you should go and change it.
Getting a card draws 10,990 federal complaints a year. Disputing something already on the statement draws 27,556. The ratio changes what you set up first.
CreditsThe credit a company with no profit was told to ignore comes off the payroll bill instead, and the election dies with the original return.
Read →The same company, on the same day, owes 85,165 dollars or 1,200. Which one depends on a box nobody tells you to tick.
Read →Cash divided by burn assumes your costs stand still. They do not, and the gap is worst for the companies that feel safest.
Read →The wire lands and a single checking account is suddenly holding two years of payroll. The layers, and the first decision is not yield.
Read →Covering five million means at least twenty banks. The mechanism is ordinary and the conditions behind it are worth an hour.
Read →One desk, working from filings and public data rather than from vendor decks. Models are published with their assumptions attached, so you can disagree with the assumption rather than with the number.