The plan and the bank statement have nothing to say to each other — until you give them one hour a month. Most small businesses write a budget once, in a burst of January energy, and never open it again. The budget was never the point. The comparison is the point: three columns, one threshold, thirty minutes, once a month. That hour is where a creeping cost is caught in month two instead of month ten.
Full checklist on the site: Budget vs Actuals — the monthly hour. Here is the whole ritual.
Three columns and one threshold
Every line gets budget, actual, variance — dollars and percent, sign included (under-budget is not automatically good; starved marketing will invoice you later). Then the rule that keeps the hour to an hour:
Any line off by more than 10% or $500, whichever is smaller, gets a written one-line cause and a one-line action. Everything else is allowed to be fine this month.
Why both measures: a $180 variance on a $2,000 line is noise; a $180 variance on a $200 line is a fire. Percent catches small-line explosions, dollars catch big-line drift.
The five lines worth reading first
- Revenue — off by 10%+? Either the plan was fiction or the market moved. Both are worth knowing in month two.
- Gross margin — the most underrated line in small business. Materials creep, subcontractor creep, and every scope increase you absorbed for free hide here.
- Payroll — usually the biggest line and the slowest to fix. It compounds monthly.
- Rent and fixed overhead — boring on purpose. These lines exist to be flat; movement is always worth the question.
- One discretionary line on rotation — travel, software, marketing. Rotate monthly; it's where budgets quietly leak.
Forty-line variance reviews die by March. Five lines read with intent survive.
The ritual — thirty minutes, same sitting as the close
- Pull actuals from the finished close, never the raw bank feed — comparing against half-closed books is how phantom variances are born.
- Mark every line past the threshold. Two to five lines is healthy. More than eight means the coding is drifting — fix the instrument before trusting the reading.
- One line per variance: cause, then action. "Fuel +22% — two country jobs; recharge travel on those jobs."
- Roll the lesson forward: a permanent change (new wage rate, new supplier price) updates the remaining months of the budget. A one-off (repair, event) does not — over-reacting to one-offs is how a budget stops matching reality within a quarter.
- Feed the 13-week cash forecast the same day: anything now known-committed becomes an outflow row.
Three honesty rules
- The budget is a decision record, not a wish. If actuals beat plan six months running, the plan was wrong in an interesting way.
- Never re-forecast the gap away. Mid-year "re-baselines" that move the budget down to meet the actuals are how a dying margin hides. Once a year, in writing, reason named.
- Classify once, consistently. Half of all variances are not performance — they're inconsistent coding that vanishes the day the same invoice always lands in the same line.
The five traps
- The built-and-shelved budget. A budget without a monthly hour is a New Year's gym membership.
- Variance theatre. Reading all forty lines aloud. Five lines read deeply beats forty read shallowly.
- Fixing the number instead of the cause. Reclassifying your way out of a payroll variance. The problem keeps drawing wages.
- "It's just timing." Sometimes true. Excused three months running, it's a trend wearing a disguise.
- Nobody owns a line. "The budget" belongs to everyone, so every variance belongs to no one. Each movable line gets one name beside it.
Worked example — the cleaning company's two un-repriced contracts
An eleven-person commercial cleaning business, $1.6M revenue, forty sites. Payroll budgeted at 46% of revenue; month two actuals landed at 53% — a $5,800 variance. The old way absorbed it as "wages went up". The new way forced a cause: two fixed-price contracts that had quietly grown — one client added two floors and a weekend service and the price never moved; the other crept from three nights a week to five across a year of polite yeses. Together they were running 38% over contract hours.
Re-quoted both at renewal: +$3,100/month, roughly $37k a year that had been leaking for an estimated fourteen months before the first review caught it. The owner's verdict: "The budget didn't find the problem. The hour did. I'd just never given it the hour."
Kits
Every page ships with a kit block — the paid tools behind the free advice:
- The First 30 Minutes — free incident quick-start checklist
- Ops Starter Kit — incident response for small teams — $14
- Ops Starter Kit Vol. 2 — advanced incident response & communications — $27
- Ops Mega Bundle — all 5 kits in one download — $49
The annual ops budget template is where the budget column comes from; the month-end close produces the actuals; the 13-week cash flow forecast is the cash-side twin; and the debtor days review is the same ritual pointed at receivables.
Top comments (0)