most finance tools tell you what already happened. closed books, last month's burn, a balance that's already old news. the number that actually changes a decision is the one 90 days out, while you can still do something about it.
fundraising is seasonal and most founders plan around it too late. start a raise in august or mid-december and you're pitching people who've mentally checked out. running low on cash during a dead window is a self-inflicted wound.
the fastest way to lose a board's confidence isn't a bad month. it's not knowing your zero-cash date when they ask. founders who can say 'we hit zero on this date, here's the plan' get trusted with a lot more rope.
every open role is a recurring cost you commit to before the revenue shows up. the real question was never 'can we afford one hire.' it's 'what's our revenue per person, and does this next hire raise it or just spend it.'
you don't need a finance team to stay alive. you need a 15-minute weekly look at burn, runway, and the next 90 days. the number that matters isn't the balance today, it's the trend it's quietly on.
'we're at 80% gross margins' sounds great until you add inference, support, and onboarding. a lot of ai companies are closer to 50. the margin you report and the margin you actually run on are rarely the same number.
if it takes 18 months to earn back what you spent acquiring a customer, you're financing growth with cash you don't have yet. cac payback under 12 months is the line between scaling and slowly drowning.
safes feel free because nothing happens at signing. but they don't remove dilution, they defer it. stack four or five and you've quietly sold a third of the company before a priced round ever sets a real number.
two companies at $1m arr can be worth 5x apart. the difference is retention. churn means you're refilling a leaky bucket every month. net revenue retention over 110% is what turns revenue into an actual asset.
paying yourself $0 isn't discipline, it's a hidden risk on your personal balance sheet. a founder one bad month from insolvency makes worse decisions, not braver ones.
the scariest place in startups: default dead with a year of runway. comfortable, still growing, and quietly not going to make it. the math doesn't care how it feels.
founders model the money coming in. almost nobody models the ownership going out. by series b most own far less than they assumed, and a cap table has no undo button.
the healthiest founder finance habit: a 15-minute weekly check of burn, runway, and the next 90 days of committed spend. not the balance, the trend. it turns "how are we doing" from a feeling into a number.
raising is seasonal. q1 and early q3 move fastest; midsummer and december stall. if your zero-cash date lands in a slow window, pull the raise forward. running out of money in august is self-inflicted.
three ways to extend runway that aren't layoffs: switch annual contracts to monthly, cut idle cloud commit, delay non-revenue hires a quarter. each buys weeks. stacked, they can buy a whole round's worth of time.
a margin founders misjudge: gross margin should include inference, support, and hosting tied to delivery. plenty of ai startups quote 80% and actually run near 50%. know your true number before you raise on it.
when competitors can call the same models you do, features stop being a moat. the durable edge is iteration speed: how fast you ship, measure, adjust. moats moved from capital, to tech, to velocity.
your bank balance tells you what already happened. what predicts survival: net burn trend, committed spend, and your real zero-cash date. track only the balance and you're driving by the rear-view mirror.
