Ecommerce Power BI Dashboard: Orders, Margin, and Returns

An ecommerce Power BI dashboard should show whether order growth is producing net sales and gross profit after discounts and refunds. Put a channel comparison beside the headline figures so the commercial team can locate the exception. Keep the source reconciliation close enough that a reader can challenge the result before acting.

Choose the summary and the detail separately

Microsoft defines a Power BI dashboard as a single page in the Power BI service, with related reports providing detail. That distinction matters when sketching an ecommerce view: the summary needs the decision signals; the supporting report needs the channel and order records that explain them.

A useful layout has a top row for scale and profit, a middle section for channel comparisons, and a bottom section for exceptions and reconciliation status. Every figure should use the same merchandise scope and cutoff date. Show when refunds are measured by posting date and when a separate view follows the original purchase cohort.

The broader Power BI dashboard examples cover other business audiences. Here the audience is an ecommerce manager deciding which channel's margin or refund activity needs attention.

Covirage is not affiliated with Microsoft.

Start with the order and adjustment exports

Mesa Outdoor, an illustrative US retailer, reviews September 2026. Its order file includes order ID, line ID, channel, merchandise value, discounts, customer ID, and completion date. A refund register identifies the original order and the reason for each adjustment. Product costs use the same units and cost policy across channels.

The table excludes sales tax and shipping revenue. Refunds include merchandise returns and customer allowances, recorded as positive deductions. COGS is the recorded merchandise cost after valid reversals for returned inventory; it is not the original cost of all units ever shipped.

Count completed orders once, even when an order has several lines. Keep canceled orders outside this count and show their adjustment treatment separately. A refund posted during September may relate to an earlier order, so this is a period reporting view rather than a lifetime view of September purchases.

Calculate what the headline cards should show

The IRS gross profit method subtracts returns and allowances to obtain net receipts, then subtracts COGS to obtain gross profit. The same relationship provides the accounting foundation for these management figures, with the scope stated explicitly.

Net merchandise sales = Gross merchandise sales − Discounts − Refunds and allowances

Gross profit = Net merchandise sales − COGS

Gross margin = Gross profit / Net merchandise sales

Direct store net sales are 72,000 − 6,000 − 6,000 = 60,000. Marketplace net sales are 48,000 − 4,000 − 8,000 = 36,000. Total gross profit is 96,000 − 54,600 = 41,400, and company margin is 41,400 / 96,000 = 43.1%, rounded to one decimal place.

Net sales per completed order divides each net sales amount by its order count. Because this period's refunds can reverse earlier purchases, label that ratio as a period measure. Do not silently substitute it for a purchase-cohort average order value with a different refund policy.

For the cost boundary and weighted total, use the separate gross profit margin guide. Payment fees, outbound fulfillment, and advertising are excluded from this example's gross profit calculation.

Give refunds their own comparison

Define the monetary refund rate here as refunds and allowances divided by merchandise sales after discounts but before refunds. This working definition keeps the denominator visible rather than relying on an ambiguous label.

Percentages are rounded to one decimal place. The company rate is 14,000 / 110,000. It measures adjustment dollars, not returned units or the share of orders receiving a refund. A physical return rate needs returned quantities and an explicitly matched quantity denominator.

The marketplace is the investigation priority: it combines the lower margin with the higher monetary refund rate. The next view should group its adjustments by product and reason, then show the underlying order references. That helps distinguish damaged goods, price concessions, and timing differences.

Change one input and see what moves

Suppose marketplace goodwill allowances fall by $4,000 while every other input stays unchanged. These particular allowances did not involve goods returning to inventory, so holding COGS constant is consistent with the scenario. This is a sensitivity calculation, not a prediction that the reduction will occur.

The revised marketplace profit is 48,000 − 4,000 − 4,000 − 21,600 = 18,400. Company profit rises by the same $4,000. The comparison shows why the adjustment reason matters before management changes a discount or returns policy.

Where it goes wrong

• Refunds counted twice. An export already showing net sales must not have the refund register deducted again.
• Percentage averages. Company margin and refund rate require summed numerators and denominators, using unrounded amounts.
• Missing costs hidden. An absent product cost is an exception, not evidence that the merchandise cost nothing.
• Summary controls promised incorrectly. Microsoft's dashboard comparison distinguishes dashboard behavior from the filtering available in reports; place detailed exploration in the appropriate view.
• Payments substituted for sales. Reconcile merchandise sales to the sales register and ledger, then explain timing and fees separately when checking settlements.

The sales identity must close: 120,000 − 10,000 − 14,000 = 96,000. The profit identity must also close: 27,000 + 14,400 = 41,400. For the wider analysis behind the display, see ecommerce business intelligence; the Power BI alternatives guide addresses different ways to answer a file-based question.

Check your ecommerce figures in Covirage

Bring your order, refund, and product-cost workbook or a Google Sheet with matching order references. Covirage maps the columns, its tools compute and check the comparisons, and the external AI model explains the changes with every figure traced to its rows. Create a free Covirage account and bring a sample of your own export.

See customer profitability analysis for the supporting business context.