Startup budgeting by department is how the strategy quietly stops mattering. Budget by priority: what are the three things that must happen this year and what does each cost? Everything else gets what is left. This forces trade-off conversations early and prevents initiative creep. Fund the three things that must happen this year first; everything else splits what is left.
Unit economics are the only financial model that matters early
Forget the five-year financial model. At early stage, the only numbers that matter are unit economics: what does it cost to acquire a customer, how much revenue does that customer generate, and how long do they stay? If customer lifetime value is greater than three times customer acquisition cost, you have a business. If it is not, nothing else matters.
Calculate CAC by dividing total sales and marketing spend by the number of new customers. Calculate LTV by multiplying average revenue per customer by gross margin by average customer lifespan in months. Both calculations should be simple enough to do on a napkin. If your CFO needs a spreadsheet, the model is too complex for your stage.
CAC payback period tells you if your growth is sustainable
Customer acquisition cost payback period is the number of months it takes for a customer's revenue to cover the cost of acquiring them. If your CAC payback is under twelve months, your growth is sustainable. If it is over eighteen months, you are spending too much to grow and will run out of cash before the customers pay back their cost.
The mistake is measuring CAC in aggregate. Break it down by channel. Your CAC from referrals might be three months while your CAC from paid ads is twenty-four months. Knowing this changes where you spend. Cut the channels with payback over eighteen months. Double down on channels under twelve. The aggregate number hides the insight.
Burn rate is a strategy as much as a number
Your burn rate should be a deliberate choice, not an accident. High burn is a bet that you will grow fast enough to raise your next round before you run out of money. Low burn is a bet that you can reach profitability or extend runway indefinitely. Neither is right or wrong. What is wrong is not knowing which bet you are making.
The burn multiple is the metric that matters: net burn divided by net new revenue. A burn multiple of one means you are spending one dollar to generate one dollar of new revenue. Below one is excellent. Above two is concerning. Above three means your growth is too expensive to sustain. Calculate it monthly and trend it over time.
Revenue recognition is simpler than you think
For early-stage B2B SaaS, revenue recognition is straightforward: recognize revenue ratably over the contract period. A twelve-month contract for twelve thousand dollars is one thousand dollars per month, recognized monthly. Annual prepayments are a liability until you deliver the service.
The mistake is booking the full contract value as revenue in month one. It inflates your numbers, misleads your investors, and creates a restatement nightmare later. Use a simple revenue recognition schedule from day one. Your future self, your investors, and your accountant will thank you.
Your financial model should fit on one page
The one-page financial model has four sections: revenue, cost of goods sold, operating expenses, and cash. Revenue is customers times average contract value. COGS is hosting plus support costs. OpEx is salaries plus tools plus marketing. Cash is beginning balance plus revenue minus expenses.
Update it monthly with actuals. Compare actuals to projections. When actuals diverge from projections by more than twenty percent, update the model. The model is not a prediction. It is a planning tool that helps you make decisions about hiring, spending, and fundraising. A model that is six months out of date is worse than no model because it creates false confidence.
Frequently asked questions
How should a startup build its budget?
By priority, not department. Name the three things that must happen this year and what each costs. Everything else gets what is left. This forces the trade-off conversation early and prevents initiative creep.
Why is departmental budgeting dangerous early?
Because every department gets its share regardless of strategy: engineering X, marketing Y, sales Z. The budget becomes an entitlement system. Priorities should eat first; departments are a reporting view, not a funding unit.
How many priorities can a startup fund at once?
Three, honestly funded. Five priorities means three are underfunded and two are fantasies. A budget that spreads evenly is a strategy document that says nothing, which is the most expensive kind.
How do I handle budget requests outside the priorities?
With the list: show what gets cut to fund the new thing. Every yes is a no to something already funded. When the trade is visible, half the requests withdraw themselves.
How often should a startup re-budget?
Quarterly, lightly. Priorities shift faster than annual cycles. The point of budgeting by priority is that reallocation is expected; an annual budget treated as scripture is where agility goes to die.