Startup finance
How to Build a Startup Budget and Financial Forecast
A startup budget is a spending plan for the next 12 months, and a forecast is your best current estimate of revenue, costs and cash. Most sources suggest a hybrid: set a top-down spending ceiling from your runway, build the detailed plan bottom-up, then run three scenarios and compare actuals against the plan every month.
Budget vs forecast: what each one is for
People mix up the two words, so it helps to separate them. A budget is a commitment: it says how much each team may spend and what revenue you aim for, and it stays fixed once approved. A forecast is an estimate that keeps changing, because it starts from what has really happened and projects forward.
Many founders keep both. The budget shows what was promised to the board or investors. The forecast shows where the company is likely to land. Pegacorn Group, a finance advisory firm, suggests freezing a copy of the approved plan and tracking actuals in a parallel copy, so the original is never overwritten. If results keep missing, a quarterly reforecast shown next to the original plan lets the board see both the commitment and the new expectation.
For an early-stage company a spreadsheet is enough. What matters is a model with clear inputs, one tab per major area and no hard-coded numbers buried in formulas. Our note on what makes a financial model good covers that in more detail.
Top-down vs bottom-up forecasting
Top-down forecasting starts from a large number, usually the size of a market, and assumes you capture a share. Bottom-up forecasting starts from inputs you control and adds them up: how many customers you can reach, what share convert, what they pay and what it costs to serve them.
| Top-down | Bottom-up | |
|---|---|---|
| Starting point | Market size or a spending cap | Customers, price, conversion, hiring |
| Typical question | “What if we win 1% of this market?” | “How many deals can two reps close?” |
| Strength | Quick sanity check on ambition | Traceable, easy to test and correct |
| Weakness | Reads as guesswork to reviewers | Slow to build, needs real data |
| Best use | Ceiling on spend, ceiling on market | Revenue and hiring plan |
The StartWise blog, written for pre-seed founders, argues that investors trust bottom-up revenue far more and that a top-down figure belongs only as a ceiling check. GrowthGrid gives a simple bottom-up example: 10,000 monthly website visitors at a 3% conversion rate and a $50 average order value produce $15,000 of monthly revenue. EY’s finance team, in a 2019 guide, suggests bottom-up for the first one to two years and top-down for years three to five.
Use both, in a set order
For spending, start top-down. Divide the cash you have by the number of months you need to last. Eagle Rock CFO gives an example of $2.0 million in cash and an 18-month target, which caps monthly burn at roughly $110,000. That figure then becomes the limit each team plans inside. For the arithmetic behind it, see our guide to burn rate and runway.
For revenue, start bottom-up and use the top-down view only to check that the result is plausible. When both methods land close to each other, confidence rises. When they diverge, the gap usually points to an assumption worth rechecking.
Step-by-step: building the budget
- Set the goals for the year. Pick a few numeric targets such as ending revenue, headcount and the month you want to reach a funding milestone.
- Split one-time costs from recurring ones. Equipment, legal setup and a website rebuild happen once, while salaries, software and rent repeat every month.
- Forecast revenue from drivers. Tie each line to something measurable, such as leads, conversion rate, price or churn, and add a ramp, because new reps and new channels take months to perform.
- Budget people carefully. Eagle Rock CFO says people costs are commonly 60 to 80 percent of startup spending and advises adding a 20 to 30 percent load on base salary for payroll taxes and benefits. Allow time for hiring too, since a role approved in March is rarely filled in March.
- Tag costs as fixed or variable. Variable costs, such as payment fees and hosting, should move with revenue in the model.
- Reconcile with the ceiling. If the bottom-up total is above the spending cap, talk through trade-offs with team leads instead of cutting every line by the same percentage.
- Keep a contingency. Several sources suggest holding back 10 to 15 percent of discretionary spending for surprises.
Marketing is a good example of a driver-based line. Eagle Rock CFO shows it as target new customers multiplied by target customer acquisition cost: 100 customers at $3,000 each means $300,000 a year, or $25,000 a month. If your pricing is still in flux, read SaaS pricing, billing and freemium before fixing revenue assumptions.
Scenarios: conservative, base and aggressive
A single forecast is one bet, while three give you a range and a plan for each outcome.
- Conservative: key growth assumptions come in lower than planned. One guide uses a miss of 20 to 30 percent as a stress test, which is a rule of thumb and not a standard.
- Base: things go roughly as planned, and this is the version you spend against.
- Aggressive: a main bet pays off sooner than expected, so you can see what extra hiring it would justify.
EY’s guide uses a similar base, worst and best case structure. Change only a few inputs between scenarios, usually conversion rate, sales cycle length, churn and hiring dates. Then write down in advance what you would do if the conservative case begins to show up, for example freezing a hire or moving a campaign to the following quarter. A trigger set in a calm month is easier to follow than one invented during a cash crunch.
Review cadence: monthly actuals, quarterly reforecast
Most sources recommend a monthly review once the books are closed. Beancount’s 2026 guide points out that forecasts are only as good as the close: if accounting takes three weeks to settle, every number you are comparing is stale.
A practical monthly pack, as described by Pegacorn Group, includes revenue, spend and net loss for the month and year to date (plan, actual, variance), spend by department, comments on any variance above a set threshold, and cash runway against plan.
Thresholds differ between sources, which is why you should pick yours once and write it down:
| Source | Variance to investigate | When to reforecast |
|---|---|---|
| Eagle Rock CFO | Over 10% or $1,000, whichever is larger | Off by 20% or more for 2 to 3 months |
| Initium Partners | Compare monthly | Off by more than 15% for two months in a row |
Rolling forecasts
A rolling forecast keeps the window the same length. Each month, actuals replace the month that just closed and a new month is added at the far end. Beancount and Eagle Rock CFO describe 12 months as the standard horizon, with some startups using 12 to 18. A common layout is one column per month, split into actual and forecast sections.
For tracking, a one-page startup financial dashboard with runway, burn and the top variances is usually enough for the founders and the board.
Common mistakes
- Building a market-share forecast and calling it bottom-up.
- Treating the annual budget as untouchable when the first two quarters have already proved the assumptions wrong.
- Leaving out payroll taxes, benefits, equipment and hiring delays.
- Reviewing results late because the books are not closed.
- Running one scenario only.
More general finance slips are listed in top finance mistakes startups make. Other topics in this area are collected in the startup finance section.
Questions people ask
Do I need a budget if I am pre-revenue? Yes. Without revenue the budget is mostly a spending plan against your runway, which is the most important number you have.
How far ahead should a startup forecast? Monthly detail for 12 to 18 months is common. GrowthGrid calls a three to five year outlook the industry standard for investors, with annual figures after the first 18 months.
How often should I update the plan? Review actuals monthly, reforecast when variances cross your threshold, and revisit the full budget quarterly.
Which is more credible to investors, top-down or bottom-up? Sources aimed at founders lean toward bottom-up, with top-down used as a reasonableness check.
Sources and limits
- Eagle Rock CFO, startup budgeting guide (eaglerockcfo.com), accessed November 2026
- Initium Partners, startup budget framework (initiumpartners.in), accessed November 2026
- GrowthGrid, financial forecast for startups (growth-grid.ai), accessed November 2026
- StartWise, top-down vs bottom-up forecasting (start-wise.io), accessed November 2026
- EY Netherlands, guide to financial modeling for startups (ey.com), 2019
- Pegacorn Group, budget vs actual for startups (pegacorngroup.com), accessed November 2026
- Beancount.io, rolling forecast vs annual budget (2026)
Most of these are advisory firms and software vendors, so percentages such as the 20 to 30 percent payroll load or the 10 to 15 percent contingency are rules of thumb, not standards. Adjust them to your own data. This guide is general information and not financial advice.