Retail analytics: questions, data and business measures

Retail analytics connects sales, stock and customer records to decisions about stores, products and channels. It explains which results changed, which rows account for the difference and what needs investigation. Begin with reconciled net sales and gross profit before adding more detailed customer or inventory measures.

What retail analytics measures

A sales increase can come from more completed purchases, a larger basket or a change in product mix. Gross profit adds the cost of the merchandise sold; stock records show whether the retailer can keep serving demand. Each answers a different business question.

Net sales = Gross merchandise sales − Discounts − Customer returns

Gross profit = Net sales − Cost of goods sold

Gross margin = Gross profit / Net sales

Net sales per completed sales transaction = Net sales / Completed sales transactions

For this article, gross merchandise sales are before separately recorded discounts and returns. Taxes collected for government agencies are excluded. The Census Bureau's retail definitions likewise treat retail sales as net of customer refunds and allowances and exclude sales tax collected for tax agencies.

The basket measure needs a policy: here, its numerator includes all returns booked during the month, while its denominator counts completed positive sales transactions and excludes voids and return-only visits. Label it accordingly rather than presenting it as the average value of newly sold baskets. For cost definitions, use the separate gross profit margin guide.

The export rows behind the measures

Start with one row per POS transaction line: date, store, channel, transaction ID, SKU, quantity, gross merchandise amount, discount and return amount. Obtain matched merchandise costs, an inventory movement export and opening and closing stock snapshots. A store and product reference list keeps renamed stores and replaced SKU codes from becoming false changes.

The retail POS analytics method covers transaction and return preparation in detail. Keep customer records separate until their keys and coverage are clear; retail customer analytics owns repeat-purchase and customer comparison questions.

Use the same closed dates, stores and channels in each comparison. Retain unmatched rows as exceptions rather than silently excluding them. A blank product cost means margin is incomplete, not that the item cost nothing.

A worked example: sales rise faster than gross profit

A fictional US home goods retailer compares August and September 2026. The exports contain these inputs, all in USD:

Subtract discounts and returns, then subtract the matched costs:

For North in September, 110,000 − 4,000 − 6,000 = 100,000 of net sales; 100,000 − 62,000 = 38,000 of gross profit. The total September check is 210,000 − 9,000 − 11,000 = 190,000 of net sales, less 120,500 of costs = 69,500 of gross profit.

Sales grew by 190,000 − 162,000 = $28,000, or 28,000 / 162,000 = 17.28%. Gross profit grew by 69,500 − 61,200 = $8,300, while margin fell by 37.78% − 36.58% = 1.20 percentage points. Percentages use unrounded calculations and are displayed to two decimal places.

North contributed $10,000 of the sales increase and South $18,000; those differences sum to $28,000. Their gross profit increases were $2,000 and $6,300, summing to $8,300. North's declining margin deserves a product-level investigation; South's margin held constant. Store totals identify where to look, but do not prove whether price, cost or mix caused the change.

Add a stock check before making a buying decision

The retailer also reconciles September merchandise quantities. For simplicity, this example has no transfers, shrink or other adjustments; units sold are net of resaleable returns.

Opening units + receipts − net units sold = closing units: 1,250 + 4,650 − 4,650 = 1,250. Reconcile by SKU before summing; a correct chain total can conceal offsetting store or product errors.

Quantity is separate from valuation. The IRS merchandise-cost guidance explains how opening inventory, purchases and ending inventory enter cost of goods sold. Do not compare a stock balance at retail ticket prices with sales costs. Seasonal styles need additional size and markdown detail, covered in fashion retail analytics.

Where it goes wrong

• Changing store coverage. Separate new and closed stores from comparisons of continuing locations.
• Averaging margins. Divide total gross profit by total net sales; store percentages have different weights.
• Repeating stock balances. A closing snapshot belongs to its date and SKU, not to every sales transaction.
• Calling a difference a cause. A store margin decline identifies a question. Product prices, costs and mix establish the explanation.

Agree on these checks before evaluating retail analytics platforms. Retail finance owns the reconciliation, while merchandising and operations own the investigation and resulting action.

Analyze your retail exports in Covirage

Bring your POS, cost and stock summaries to Covirage; its tools compute and check the measures, and the external AI model explains the results with every figure traced to its rows. Create a free Covirage account, which takes about a minute, then set up your business and bring a sample of your own data.

See self-service analytics for the supporting workflow.