Most small businesses have a budget in the sense that the owner has a feel for what things cost. Fewer have a budget in the sense that matters: a set of decisions, written down in advance, about what the business will earn and spend, that gets compared with reality every month.
The difference between those two is the difference between noticing a problem when the money runs out and noticing it in the first week it starts. So let's build the second kind, step by step, and then look at how to use it, because a budget that gets written once and filed away does very little.
Three numbers that get confused
It helps to separate three things that often get called "the budget."
The budget is your plan. It's what you decided, at the start of the year or quarter, the business should earn and spend. It stays fixed, so you have something to measure against.
The forecast is your current best estimate of what will actually happen, updated as you learn more. It moves.
The actuals are what did happen, from your books and bank statements.
The useful information lives in the gaps between them. The gap between budget and actual, called the variance, tells you where reality departed from your plan. The gap between budget and forecast tells you where it's about to.
It's worth seeing why a business might run without one, and what that costs. A cash surprise is usually the result of spending nobody compared with a plan. That's the result of having no plan to compare it with. And having no plan is often the result of a belief that budgets are paperwork for accountants. Each step follows from the one before, and the first one is easy to change.
Step 1: Start with sales, built from drivers
Every other line in a budget depends on sales, so start there, and build it from the things that cause sales instead of from a growth percentage.
For a shop, that might be footfall × the share who buy × average spend. For a service business, it might be number of clients × hours per client × rate. For a distributor, it might be number of accounts × average monthly order.
Do it month by month, because almost every business in Sri Lanka has a seasonal pattern: festival seasons, school terms, tourist seasons, the year-end. Use last year's monthly sales as the starting shape and adjust for what you know is changing.
Step 2: Direct costs as a share of sales
Direct costs are the ones that rise and fall with sales: the goods you resell, raw materials, packaging, sales commissions, delivery per order. Set them as a percentage of sales, based on your recent history. If your goods have been costing about 60% of what you sell them for, budget 60%, and adjust if you already know a supplier price rise is coming.
Step 3: Overheads, line by line
Overheads are the costs that stay roughly the same whatever you sell: rent, salaries, utilities, software, insurance, accounting, marketing, loan repayments.
Two things often go missing here.
The first is the full cost of staff. An employee costs more than their salary: on top of gross pay, the employer contributes 12% to the EPF and 3% to the ETF. So a monthly payroll of LKR 380,000 costs the business LKR 437,000.
The second is anything that doesn't come round every month: annual licences, insurance renewals, festival bonuses, equipment servicing, tax payments. Put each one in the month it falls due, or set aside a monthly amount for it, so it never arrives as a surprise.
Step 4: Put it together and look at the result
A simple monthly budget for a small retailer might look like this:
| Sales | LKR 2,000,000 |
| Direct costs (60%) | LKR 1,200,000 |
| Gross profit | LKR 800,000 |
Staff (pay 380,000 + EPF 45,600 + ETF 11,400) LKR 437,000
Rent LKR 120,000
Utilities LKR 45,000
Marketing LKR 60,000
Other (software, accounting, transport) LKR 40,000
Total overheads LKR 702,000
Net profit before tax LKR 98,000
Now step back and ask whether this is a business you want to run. A 4.9% net margin (the share of sales left after every cost) leaves little room for a bad month. That's the first use of a budget: it shows you the consequences of the current plan while you can still change the plan. If the result is too thin, you can look for higher prices, better supplier terms, more sales from the same overheads, or overheads that aren't earning their place.
Step 5: Compare, every month
This is the step that makes a budget useful, and it takes about an hour a month.
When the month closes, put the actuals next to the budget, line by line. Say the retailer's month came in like this:
Sales: LKR 1,850,000 against a budget of 2,000,000 (7.5% under).
Direct costs: LKR 1,147,000, which is 62% of sales against a budget of 60%.
Utilities: LKR 58,000 against a budget of 45,000.
Everything else on budget.
Gross profit (sales minus the direct cost of what was sold) comes to LKR 703,000 and overheads to LKR 715,000. The planned profit of LKR 98,000 has become a loss of LKR 12,000.
Then ask three questions of each line that moved.
What happened? Sales were lower; goods cost more as a share of sales; electricity jumped.
Is it a one-off or a pattern? A slow month after a holiday is different from a third slow month in a row. A one-time repair is different from a tariff change.
What will we do about it? Maybe nothing, if it was a one-off. Maybe a price review, if supplier costs have risen for good. Maybe a closer look at the electricity bill.
The value is in the second question. One variance is information. The same variance three months running is a trend, and a budget compared monthly lets you see the trend in month two, while it's still small.
Cutting costs and controlling costs are different activities
When money gets tight, the reflex is to cut. Cutting has its place, but it's worth distinguishing it from controlling.
Cutting is a one-time reduction. It's fast, and it can do damage, because every cost is connected to something. Cut the marketing budget and enquiries fall a month later. Cut a salesperson and the sales they were bringing in go with them. Switch to a cheaper supplier and quality complaints rise. The saving shows up immediately; the cost of it shows up later, somewhere else, and is often larger.
Controlling is an ongoing habit. It means knowing what each cost is for, what it produces, and whether it's still worth it. A few practices that help:
Review every recurring payment once a year. Subscriptions, service contracts and standing orders tend to outlive the need for them.
Track costs per unit of output as well as in total. Electricity per unit produced, delivery cost per order, marketing cost per new customer. A total that rises because you sold more is fine. A cost per unit that rises is a signal.
Ask what each cost produces before cutting it. Some costs produce revenue. Some protect the business. Some do neither. The third group is where to cut first.
Set approval limits. Decide in advance who can spend how much without checking with you. It keeps small purchases moving and puts a pause in front of large ones.
Three attitudes to budgets
The first is that a budget is a box to tick for the bank. The second, after a year when the budget was wrong by March, is that budgets are pointless because things always change. The third treats the budget as a prediction you deliberately compare with reality every month, so that being wrong becomes the most useful part of it.
Does that make sense? A budget earns its keep by being checked, the way a map earns its keep by being held up against the road.
Budgets as a way of seeing ahead
A budget can feel like paperwork, something the bank or the accountant asks for. It's more useful to think of it as a set of predictions you've made about your own business, which you then check against what happens. Every month you compare, you learn where your predictions were wrong, and your picture of how the business actually works gets more accurate. After a year, you'll know your seasonal pattern, your real cost structure and your margins (the share of each sale left after paying for the goods) far better than you did when you started. And the next budget you write will be much closer to the truth.
How Gridwerk helps
Building a budget from your real cost structure, setting cost and approval policies, and running the monthly review with you are part of our Financial Intelligence, Forecasting and Process & Financial Control work. If you don't yet have a clear monthly view of what your business earns and spends, start with the free diagnostic. It's eighteen questions, about seven minutes, and then we call you to go through the rest.
Gridwerk diagnostic