Cost of goods sold looks like a definitional question and behaves like a strategic one. Move one cost across the line and your gross margin changes without anything in the business changing at all.
Cost of goods sold is the direct cost of what you actually sold in a period. Not what you bought, not what is sitting in the stockroom — what left. That distinction is the whole of the calculation, and it is why the formula involves opening and closing stock rather than a simple total of purchases.
The number matters because everything above it and below it is judged against it. Gross margin is revenue minus this figure; profitability comparisons across periods are meaningless if the contents of the line shift. And they do shift, quietly, whenever somebody decides that delivery or packaging belongs somewhere else.
For a service business the equivalent is cost of sales: the labour delivering the work, plus anything bought specifically for it. The mechanics differ, the principle does not — direct costs of what was delivered, and nothing else.
The consistency matters more than the classification. Reasonable businesses disagree about whether outbound delivery sits in cost of goods sold or in operating expenses, and either answer works — as long as it is the same answer every month. A business that reclassifies mid-year produces a margin change that looks like a commercial trend and is an accounting decision, and that misreading has led people to celebrate or panic about nothing.
Divide it into revenue and you have gross margin, which is the number that decides what the business can afford — covered in gross margin. Track it per product line and the picture usually changes: most businesses have a line they sell enthusiastically at a cost they have never actually calculated. And watch the ratio of stock to cost of goods sold over time, because rising stock against flat sales is working capital quietly disappearing into the stockroom, which is the ground covered in working capital.
Ettex Books produces the figure from purchases and stock movements, provided both are recorded; the stock side sits in inventory management, and the wider statement it feeds into is profit and loss statement.
The limit worth naming: no software can decide what belongs in your cost of goods sold. That is a policy choice, and the value of making it explicitly is that next year's comparison means something.
Opening stock plus purchases minus closing stock. Purchases include inbound freight and duty; the stock figures must be counted on the same basis at both ends.
Inbound shipping, yes — it is part of what the goods cost you. Outbound delivery to customers is treated either way by different businesses; pick one and stay consistent.
Usually called cost of sales: the labour directly delivering the work plus anything bought specifically for the job. Administrative salaries do not belong there.
Most often because something moved across the cost of goods sold line, or because stock was counted differently. Both look like commercial trends and are accounting decisions.
Statutory accounts are for the tax authority and arrive months late. Management accounts are for you, arrive within two weeks, and are worth more for being approximate and on time than exact and irrelevant.
Burn rate is the simplest number a business can track and the one most avoided. Divided into your cash balance it produces runway — the number of months before a decision is made for you.
Gross margin is not markup, and confusing the two is the most expensive arithmetic error in small business. A 50% markup is a 33% margin — and the shortfall shows up as a hole in the fixed costs.