How to Create a Business Budget That Actually Guides Decisions

Most owners who ask how to create a business budget end up with the same artifact: a spreadsheet that is accurate in January, stale by March, and never opened again after the second quarter. The problem is almost never the math. It is that the budget was built as a document to have rather than a tool that answers two specific questions — can I afford this hire and will I have cash in month seven.

There are really only two structures worth building at the small-business level, and they behave very differently. This is a head-to-head between them: the static annual budget and the rolling twelve-month forecast. Both start from the same place — your own profit and loss statement — and both are legitimate. They just fail in opposite directions, and picking the wrong one is why the spreadsheet dies in March.

Two Ways to Build a Budget From the Same P&L

A static annual budget is set once, usually in November or December, and locked. Twelve columns, one per month, agreed to before the year starts. You measure actual against plan each month and explain the variance. Nothing moves.

A rolling twelve-month forecast always looks twelve months forward from wherever you are standing. When August closes, you drop August, add next August, and update the numbers you now know that you did not know in December. The horizon never shortens.

The distinction sounds academic until the first time reality diverges from the plan. In a static budget, a lost anchor client in April means every remaining month is now measured against a number that no longer means anything. In a rolling forecast, that same event forces a rebuild of the next twelve months on the spot — which is the moment the budget stops being a report and starts being a decision.

Not sure what your books actually need? Get an instant estimate in about two minutes.

How Do You Create a Business Budget From Last Year's P&L?

Both structures share the same first four steps. Do these before you decide which shape to commit to.

1. Export twelve months of actuals by month. In QuickBooks Online, that is Reports > Profit and Loss, with the period set to last twelve months and the columns set to Months. Not a summary. You need the month-by-month grid, because the whole point is finding the shape of the year, not the average. If those twelve months are not trustworthy — uncategorized transactions, unreconciled accounts, personal spending mixed in — stop and fix that first. Budgeting on top of bad books produces a confident wrong answer. Our run-today catch-up audit is the faster path if you are behind.
2. Split every expense line into fixed, variable, or step. Fixed costs do not move with volume: rent, software subscriptions, insurance, base salaries. Variable costs move roughly in proportion to revenue: materials, subcontractors, card processing, hourly field labor, fuel. Step costs are the ones owners forget — they stay flat, then jump all at once. A second truck. A shop lease. A supervisor. Step costs are the entire reason budgets break.
3. Convert the variable lines to percentages of revenue. If materials ran $312,000 on $1.1M of revenue, you budget materials at 28.4% of whatever revenue you plug in — not as a flat $26,000 line repeated twelve times. This one move is what makes a budget survive a bad quarter, because the cost side moves when the revenue side does.
4. Rebuild the revenue line from units, not from a growth percentage. "Fifteen percent growth" is a wish. "Forty-two recurring accounts at an average monthly value of $2,150, plus roughly nine project jobs a quarter" is a forecast you can check monthly against reality. If you cannot state your revenue as a count times a price, you do not have a revenue forecast — you have a target.

Step four is where the two structures diverge. Read your last twelve months honestly before you fill in the next twelve; if the P&L grid is not something you read fluently yet, start with how to read a profit and loss statement, because everything below assumes you can spot a seasonal dip from a collections problem.

Annual Budget vs. Rolling Forecast: The Comparison

That last row is a practical constraint worth knowing before you start. QuickBooks Online's built-in Budgets feature is limited to Plus and Advanced plans, it budgets by account for one fiscal year at a time, and it has no native rolling window. It is genuinely good at what it does — importing your prior-year actuals and running a Budget vs. Actuals report — but it cannot roll. If you choose the rolling forecast, you will export to a spreadsheet and keep QBO as the source of actuals.

Which Business Budget Method Should You Pick?

Pick the static annual budget when: your revenue is 70%+ recurring or contracted, headcount is stable, you are not planning a truck, a lease, or a hire this year, and you need a fixed number to hold a manager or yourself accountable to. Stability is the qualifier. A cleaning company with 40 recurring commercial accounts and two crews does not need a rolling forecast; it needs a plan it does not renegotiate every time a month comes in light.

Pick the rolling twelve-month forecast when: any single client is more than 15% of revenue, your work is project-based with uneven timing, you are planning to add headcount, or you have added or lost a meaningful revenue line in the last six months. Growth and lumpiness both break a static plan, and they break it in the same way — the plan stops describing the business somewhere around month four.

If you are genuinely on the fence, run the rolling forecast and freeze a copy of month one as your annual plan. You keep the accountability of a fixed number and the decision-making value of a live one, at the cost of about an hour a month of your own time. That is the setup I recommend most often to owners doing somewhere between three-quarters of a million and three million in annual revenue, because that is exactly the band where businesses hire their first non-owner manager and the static budget stops keeping up.

What Breaks a Business Budget by March?

In almost every abandoned budget I have looked at, the failure is one of four things — and all four are structural, not arithmetic.

• Payroll taxes were spread evenly across twelve months. They are not even. The FUTA taxable wage base is $7,000 per employee, and state unemployment wage bases are similarly capped, which means federal and state unemployment tax for a stable staff is largely spent by the end of Q1. Divide annual payroll tax by twelve and your January through March will look like a disaster and your fourth quarter will look artificially strong.
• Biweekly payroll was budgeted as "twice a month." Biweekly is 26 pay periods, not 24. Two months of every year carry three paydays. If you budget payroll as annual divided by twelve, those two months blow the plan by roughly 50% of a payroll — and the owner concludes the budget is broken rather than the assumption.
• Estimated taxes were treated as quarterly and equal. The IRS estimated tax due dates are April 15, June 15, September 15, and January 15 — unevenly spaced, and the January payment lands in a different calendar year than the income it covers. What you actually owe on each date depends on your entity, your owner compensation, and your state; that is a conversation for your CPA, but the dates belong in the budget as four specific cash outflows, not as a monthly reserve line you never fund.
• The budget was built on revenue instead of on cash. A budget that says you will be profitable does not tell you whether you can make payroll on the 15th. If you invoice net-30 and collect in 41 days, a profitable March is an empty April. This is the most common failure of the four, and it is worth understanding on its own terms — profit and cash flow are different measurements, and a budget that only models the P&L answers only half the question.

Fix all four and add one column to the right of your monthly plan: projected ending cash. That column is the reason the budget exists.

How Do You Budget for a Hire Without Guessing? A Worked Example

Take a commercial cleaning company doing $1.6M a year, roughly $133K in monthly revenue, with a dip of about 12% in July and December when office buildings empty out. The owner wants to add a $62,000 operations manager in May.

The static budget answer is a single line: $62,000 divided by twelve, added to overhead from May forward. It says the business can absorb it. Here is what the rolling forecast shows instead.

• Fully loaded cost, not salary. $62,000 base, plus roughly 7.65% employer FICA ($4,743), plus federal and state unemployment concentrated in the first quarter of employment, plus a health contribution, plus a phone and a laptop. The real number lands closer to $74,000-$78,000 annually — call it $6,400 of added monthly cost from May forward.
• Front-loaded, not smooth. Because the unemployment wage bases reset per employee, May and June carry the SUTA and FUTA cost for that hire almost entirely. Months one and two are the expensive ones.
• The trough is July, not May. May and June absorb the new cost against full revenue. July arrives with the seasonal 12% dip — about $16,000 less revenue — while carrying the new $6,400. The rolling forecast shows ending cash bottoming out in late July at a number the owner needs to look at before signing an offer letter, not after.
• The offset is not immediate. The manager is hired to free up 15 owner hours a week. That converts to revenue only if those hours go into selling and the pipeline supports it. Model the revenue lift starting in month three or four, not month one, and model it as a count of new accounts rather than as a percentage.

The decision that comes out of this is usually not "no." It is "yes, but hire in March instead of May, so the payroll tax front-load and the ramp period both clear before the July dip." That is what a budget is supposed to produce — a changed decision, with a date attached. Pair it with a rolling thirteen-week view of actual bank cash, which is covered in more depth in our guide to cash flow management.

How Often Should You Update a Business Budget?

Monthly, within five business days of the books closing. Not quarterly. The reason is timing: a variance you catch in the first week of the following month is still actionable, and a variance you catch 90 days later is history. According to the U.S. Bureau of Labor Statistics, about one in five new establishments closes within its first year and roughly half are gone by year five — and the businesses that fail rarely do so because they lacked a budget. They fail because nobody looked at the cash position early enough to change course while changing course was still cheap.

A workable monthly cadence takes under an hour: close the books, pull Budget vs. Actuals, investigate any line more than 10% or $1,000 off plan (whichever is larger), roll the forecast forward one month, and write two sentences about what you are changing. That last step is what separates a budget from a report. If you want a fuller picture of what belongs in the monthly review, see which financial reports to read and when.

So Which Business Budget Should You Build This Year?

If your revenue is stable and recurring and you are not adding people this year, build the static annual budget in QuickBooks Online and review variances monthly. If you are hiring, growing, or dependent on a handful of clients, build the rolling twelve-month forecast in a spreadsheet, freeze month one as your annual plan, and rebuild it every month with a projected ending cash column on the right.

Either way, the budget only works if the underlying books are current and categorized consistently — a forecast built on a P&L with three months of uncategorized transactions is a well-formatted guess. Get the twelve months of actuals right first. The structure is the easy part.

Frequently asked questions

Should I budget by month or by quarter?

By month. Quarterly budgeting hides the two things that actually cause cash problems: three-paycheck months on a biweekly schedule and seasonal revenue dips that only last four to six weeks. A quarter can look perfectly on plan while one month inside it ran the operating account down to nothing.

Can I just use the QuickBooks Online budget feature?

Yes, for a static annual budget. QBO Budgets is available on Plus and Advanced, it will pre-fill from your prior-year actuals, and its Budget vs. Actuals report is useful for monthly variance review. Its limits are that it works one fiscal year at a time and cannot roll forward, so a rolling twelve-month view means exporting the P&L into a spreadsheet.

What percentage of revenue should payroll be?

There is no universal number, and any figure quoted without your industry attached is not worth much. What matters more is the trend in your own business: pull payroll as a percentage of revenue for each of the last twelve months and look at whether it is drifting up. A service business where that percentage climbs three months in a row is either underpricing or overstaffed.

How do I budget for taxes if my income is unpredictable?

Budget the cash outflow on the four estimated tax dates rather than trying to predict the exact amount, and move a set percentage of every deposit into a separate tax savings account so the money exists when the date arrives. The correct percentage depends on your entity type, owner compensation, and state — that is a question for your CPA, ideally before the year starts.

My budget was wrong within two months. Should I start over?

No. Being wrong is the normal state of a forecast; the useful question is which assumption was wrong. If revenue missed, your unit count or pricing assumption was off. If costs missed on a percentage line, your variable rate was off. If costs missed on a fixed line, something structural changed. Fix the specific assumption and roll forward.