Blog · Board and management reporting
Business intelligence for ecommerce explains why sales, returns and margin change. This guide defines the measures, works a monthly example, and assigns the checks and decisions to the people who own them.
Business intelligence for ecommerce explains how orders, discounts, returns and costs produce the revenue and margin a company actually retains. Its value comes from connecting a checked result to a business decision and the transactions behind it.
Start with questions such as whether sales growth survived refunds, whether product mix changed gross profit, and whether a channel covers its assigned variable costs. The broader business intelligence definition provides the general context; the ecommerce work begins with the company's own commerce exports.
Decide which orders count as ecommerce before comparing periods or channels. The US Census Bureau's retail definitions identify ecommerce by how the order is placed or its terms are negotiated; payment does not have to occur online. A cash-on-delivery online order can therefore belong in the commercial scope even though its payment arrives elsewhere.
For management reporting, document included sales channels, order statuses, currency, dates and tax treatment. Finance must also explain how the operational view ties to recognized ledger revenue. An order-created report and a shipment-based ledger can differ legitimately, but an unexplained gap weakens every conclusion built on them.
Use a small set of measures with explicit meanings:
| Measure | Definition used here | Decision it supports |
|---|---|---|
| Gross merchandise sales | Product price multiplied by sold quantity, before discounts and merchandise reversals | Assess the initial sales activity |
| Discounts | Allocated reductions from the merchandise selling price | Review promotion and pricing decisions |
| Merchandise refunds | Net merchandise revenue reversed after the original discount | Investigate refund burden and reasons |
| Net merchandise sales | Gross sales less discounts and merchandise refunds | Compare retained merchandise revenue |
| Gross profit | Net sales less net product cost | Investigate product economics and mix |
| Contribution | Gross profit less the specified variable selling costs | Review the remaining amount toward fixed costs |
Keep taxes, customer shipping revenue and their reversals separate in this example. Shopify's sales-report documentation distinguishes net sales from total sales and financial reversals from physical returns. Match the definition behind an export column to the business question before reusing its label. Covirage is not affiliated with Shopify.
An illustrative US home-goods company compares August and September 2026. Finance supplies net product costs after approved cost recoveries. Variable selling costs include fulfillment, transaction fees and assigned marketing expenses; corporate overhead stays outside contribution.
The company uses processed-date sales adjustments for both months. These figures describe period activity, including any adjustments processed against older orders. They do not measure the eventual return rate of each month's new orders.
| Export total or computed measure | August (USD) | September (USD) | Calculation |
|---|---|---|---|
| Gross merchandise sales | 100,000 | 120,000 | Supplied order-line totals |
| Discounts | 5,000 | 15,000 | Supplied allocated discounts |
| Merchandise refunds | 5,000 | 15,000 | Supplied refund-line totals |
| Net merchandise sales | 90,000 | 90,000 | Gross sales − discounts − refunds |
| Net product cost | 45,000 | 54,000 | Supplied cost and recovery totals |
| Gross profit | 45,000 | 36,000 | Net sales − net product cost |
| Variable selling costs | 18,000 | 22,000 | Supplied expense totals |
| Contribution | 27,000 | 14,000 | Gross profit − variable selling costs |
September gross merchandise sales rose by $20,000, or 20.0% of August's $100,000. But the additional $10,000 of discounts and $10,000 of merchandise refunds absorbed that entire increase. Net merchandise sales remained $90,000.
Gross margin is gross profit divided by net sales: $45,000 divided by $90,000 gives 50.0% in August; $36,000 divided by $90,000 gives 40.0% in September. Contribution margin falls from 30.0% to 15.6%. Percentages are displayed to one decimal place, using unrounded calculations.
The increase in gross sales is real within the agreed scope. It is insufficient evidence that the business improved. The ecommerce reporting guide explains how to prepare the underlying product and channel package without duplicating transactions.
Start with August contribution, then apply the changes in the source rows:
| Contribution bridge | Change (USD) | Running contribution (USD) |
|---|---|---|
| August contribution | 27,000 | |
| Higher gross sales | 20,000 | 47,000 |
| Additional discounts | -10,000 | 37,000 |
| Additional merchandise refunds | -10,000 | 27,000 |
| Higher net product cost | -9,000 | 18,000 |
| Higher variable selling costs | -4,000 | 14,000 |
| September contribution | 14,000 |
The bridge closes to the September result. Its $13,000 decline consists of $9,000 of additional product cost and $4,000 of additional variable selling costs after the net-sales effects cancel.
This is an arithmetic explanation, not proof of a root cause. Higher product cost could reflect mix, supplier cost, recoverability of returns or timing. To choose among those explanations, compare matched product and order rows and inspect the cost policy. The business intelligence charts guide covers choosing a display once the measure and bridge are established.
The trading manager owns the discount investigation: which promotions and products account for the increase, and did those orders retain enough margin? Customer operations owns the refund investigation: which reasons and original order cohorts explain the reversals?
Finance owns the cost reconciliation. Missing costs, changed costing assumptions and returned goods written down rather than restocked need resolution before the team attributes the gross-profit decline to pricing. Marketing and operations own the variable-cost detail and any proposed spending changes.
Record the evidence needed, the action owner and the next review date alongside the finding. Carry material exceptions into management reporting so the same result reaches the people who can approve changes. The broader ecommerce business intelligence guide connects this question-based analysis to the wider commerce reporting context.
Bring a sample of your own order, refund and cost files; Covirage maps the columns, and its tools compute and check the revenue and margin figures. The external AI model explains the results with every figure traced to its rows. Create a free Covirage account and set up your business.
See revenue driver analysis for the supporting commercial questions.
Its purpose is to explain commercial performance well enough to make decisions. A finance or trading team can examine discounts, merchandise refunds, product costs and channel expenses together, then investigate which products, orders or channels account for a change.
Start with order lines, merchandise refund details and product costs for the reporting period. Add fulfillment, transaction-fee and marketing exports when measuring contribution. Preserve order IDs, line IDs, product codes, channels, dates and currency so adjustments can be matched and totals checked.
Finance should approve revenue and cost definitions and their reconciliation. Ecommerce trading owns merchandising and discount actions; customer operations investigates return reasons; marketing owns acquisition spending. Name a person for each exception and an agreed date to review the evidence.