Retail & CX metrics glossary
COGS is the cost of the merchandise you actually sold. Here is how to calculate it — and where the margin it plans gets lost.
Cost of goods sold (COGS) is the direct cost of the merchandise a retailer sold in a period: what the goods cost to buy, plus inbound freight and non-recoverable taxes on the purchase. It deliberately excludes everything needed to run the store — rent, payroll, marketing, utilities — which lives below the line as operating expense. The boundary matters because COGS defines gross margin: revenue minus COGS is the money left to pay for the entire operation and, hopefully, profit.
COGS is where retail profitability is planned. Every negotiation with a supplier, every shrinkage percentage point, every shift in the product mix lands here before it lands anywhere else. This page covers the periodic formula used in practice, a working calculator, honest benchmarks of COGS as a share of revenue by segment — and the blind spot that makes an excellent COGS worthless when the sales floor doesn't convert.
COGS — Cost of goods sold
COGS = beginning inventory + purchases − ending inventory
beginning inventory = stock at cost at the start of the period · purchases = merchandise bought in the period, including inbound freight, net of returns and discounts · ending inventory = stock at cost at the end of the period, from a physical count or perpetual records
COGS calculator
A store starts the quarter with $180,000 of inventory at cost, buys $620,000 of merchandise, and ends with $200,000 on hand. COGS = 180,000 + 620,000 − 200,000 = $600,000. If revenue for the quarter was $1,000,000, COGS is 60% of sales and gross margin is 40% — the entire operation, from rent to payroll, has to fit inside that 40 cents per dollar.
What is a typical COGS as a percentage of revenue?
COGS ratios are largely set by the category's economics — volume businesses run high, margin businesses run low. As a rough map from public industry aggregates:
| Grocery & supermarkets | 70–80% of revenue |
|---|---|
| Consumer electronics | 70–80% of revenue |
| Pharmacy & drugstores | 65–75% of revenue |
| Fashion & apparel | 40–55% of revenue |
| Restaurants & food service | 28–38% of revenue |
Ranges compiled from public financial aggregates of listed companies — treat as orientation, not targets; private-label share, tax regime and how shrinkage is booked shift the ratio between otherwise similar retailers.
How to calculate COGS in practice: inventory plus purchases
Most retailers compute COGS by the periodic method shown above: whatever was here, plus whatever arrived, minus whatever is still here, must have been sold — or lost. That last clause is the catch. The formula cannot tell a sale from a theft, a breakage or a counting error: unrecorded shrinkage silently inflates COGS and quietly deflates gross margin. This is why inventory accuracy and a separately measured shrinkage line are prerequisites for a COGS you can trust.
Retailers with perpetual inventory systems recognize cost at the moment of each sale instead of waiting for a count, which gives a live COGS by SKU, category and store. The periodic count then becomes an audit that reconciles the system to reality. Whichever method you use, consistency in valuation (cost method, treatment of freight and supplier rebates) matters more than the method itself — changing the rules mid-year makes every comparison meaningless.
How to reduce COGS without killing the mix
The classic levers: negotiate cost and rebates with suppliers on the strength of your sell-out data, attack shrinkage as an operational program rather than an accepted loss, recover every recoverable tax the regime allows, and steer the product mix toward items with better cost-to-price ratios — including private label where the brand permits. Each point of COGS recovered drops straight into gross margin with no extra sales effort.
The trap is cutting cost in ways that shrink revenue faster than they shrink COGS: gutting assortment variety, downgrading quality the customer notices, or concentrating purchases in one cheap supplier until stockouts appear. COGS is a ratio with revenue underneath it — a decision that saves 2% of cost but drives customers elsewhere makes the percentage worse, not better. Mix decisions should be judged on total gross margin produced, not on the cost line alone.
The blind spot: planned margin vs realized margin
COGS defines the margin you planned; the sales floor decides the margin you get. An excellent buying operation that lands merchandise at a great cost still doesn't pay the store if the conversion is weak, if higher-margin items are never offered, or if discounts are handed out at the counter to close hesitant sales. The spreadsheet says 45% gross margin; the till says otherwise.
Between the two sits the sales conversation — the least measured link in retail. Which product the salesperson chooses to show first, whether an attach is suggested, how a price objection is handled: those decisions move realized margin every hour of every day, and none of them appear in the COGS report. Margin is planned in COGS and realized in the conversation; managing only the first half is managing half the P&L.
The income statement shows the cost. We show why the margin doesn't land.
Cognifyze captures the in-person sales interaction itself — with consent, without identifying any individual shopper — and connects the margin you planned in COGS to what actually happens at the moment of sale: what gets offered, what gets discounted, which high-margin item never leaves the shelf because nobody mentions it. A census of interactions, not a sample of opinions.
In measured deployments, making the interaction visible moved same-store conversion from 51.5% to 79.5% (+28pp, p<0.001), with 383% ROI and payback in 1.4 months.
See where your planned margin leaks on the floor — book an executive diagnostic.
Related metrics and guides
COGS — frequently asked questions
What is included in COGS — and what is not?
Included: the purchase cost of the merchandise sold, inbound freight to get it to you, and non-recoverable taxes on the purchase, net of supplier returns and discounts. Excluded: rent, salaries, marketing, utilities, outbound delivery and every other cost of running the operation — those are operating expenses, below gross margin.
What is the difference between COGS and operating expenses?
COGS varies with what you sell — sell nothing and it is zero; operating expenses exist whether you sell or not. The distinction anchors the income statement: revenue minus COGS gives gross margin, and gross margin minus operating expenses gives operating profit. Mixing the two hides where a profitability problem actually lives.
How does COGS relate to gross margin and markup?
Gross margin is revenue minus COGS, expressed in dollars or as a percentage of revenue; markup is the same gap expressed as a percentage of cost. A product bought at $60 and sold at $100 has a 40% gross margin but a 66.7% markup. COGS is the shared foundation — get it wrong and both numbers are fiction.
Is shrinkage part of COGS?
By default, yes — anything that left inventory without being at the start or the end of the period falls into the periodic formula, sold or not. Well-run retailers measure shrinkage separately (cycle counts, a dedicated loss line) precisely so that COGS reflects genuine sales cost and the loss problem stays visible instead of dissolving into margin.
Is a lower COGS always better?
Only if revenue holds. Cutting cost by degrading quality, narrowing the mix or squeezing a single supplier can push customers away faster than it saves money — and since COGS is judged as a share of revenue, falling sales can make the ratio worse even as absolute cost drops. Judge cost decisions by the total gross margin they produce.