Protect or replenish
Strong relevant movement with limited availability may justify buying review, subject to lead time, incoming stock and the future selling window.
Practical analytical guidance
SKU Performance Analysis
A practical guide to evaluating SKU sales, sell-through, stock cover and inventory investment—and translating product signals into defensible decisions.
Begin with a decision, not a best-seller list.
SKU performance analysis evaluates individual stock-keeping units in relation to sales, available stock, commercial role and the decision being made. The same SKU may be strong on unit movement but weak on margin, or low on sales because it was unavailable. A revenue ranking alone cannot resolve replenishment, markdown or assortment choices.
Use a defined question: which products need buying attention this month, which items need a range review, or where inventory exposure is concentrating. Choose a consistent review period and compare products with relevant peers. A launch-week item should not automatically be assessed against a product that has been available for six months.
Build a consistent product-level view.
- Identify the grain. Confirm whether each record is an order, an order line, a daily SKU total or an inventory snapshot. Keep a stable SKU key and a clear variant-to-style-to-category hierarchy.
- Define sales. Separate units, gross sales, discounts, returns and net sales. Decide how cancellations and later refunds are treated. Never sum an order total repeatedly across its line items.
- Align the period. Use matching start and end dates for sales comparisons and state the inventory snapshot date. Record launch dates, promotion periods and unavailable days when known.
- Check the reconciliation. SKU totals should roll up to category totals and the agreed source totals. Investigate duplicate identifiers, unmapped products and unexplained negative quantities before ranking performance.
The inventory analysis data checklist sets out useful export fields and validation questions.
Read several measures together.
| Measure | What it helps clarify | Interpretation limit |
|---|---|---|
| Units and net sales | Movement and commercial contribution | Price, availability, returns and mix can change the comparison. |
| Average weekly unit sales | Recent movement over a stated number of weeks | A full-period average can understate unconstrained demand when stock was unavailable. |
| Sell-through | Progress against a defined quantity available for sale | Use a reconciled opening-stock/receipt basis or a clearly defined cohort; do not mix denominators. |
| Stock cover | Units on hand relative to weekly unit sales | Historical cover is not a demand forecast; no recent sales means cover is undefined, not zero. |
| Inventory at cost | Capital held in the product | Requires a consistent cost basis and does not measure market value. |
| Gross-margin contribution | Sales contribution after applicable product cost | Requires compatible net-sales and cost-of-goods data; it is not net profit. |
Two SKUs, different inventory positions and buying priorities.
Illustrative example with fictional data
Assume both products were continuously available for an eight-week period. SKU A sold 80 units and has 30 units on hand. SKU B sold 40 units and has 120 units on hand. Ignore incoming orders for this first comparison.
Average weekly unit sales = units sold ÷ weeks in the review period
Stock cover in weeks = units on hand ÷ average weekly unit sales
SKU A: 80 ÷ 8 = 10 units per week; 30 ÷ 10 = 3 weeks of cover. SKU B: 40 ÷ 8 = 5 units per week; 120 ÷ 5 = 24 weeks of cover.
A deserves an availability review before the next supply decision, especially if its lead time exceeds current cover. B deserves an inventory-commitment review. Neither observation alone establishes an order quantity or a markdown: incoming stock, future selling windows, margin, minimum order quantities and product role still matter.
For a seasonal item, the remaining selling window may matter more than a trailing average. For an intermittent seller, several zero-sales weeks may be normal. If stockout days are known, compare observed sales with in-stock sales velocity and disclose the difference. Do not relabel observed sales as total customer demand.
Translate the evidence into decision options.
Monitor or investigate
Unclear trends, recent launches, promotions or incomplete availability history may justify another observation period or a targeted data check.
Reduce or rebalance
High stock exposure with weak movement may warrant a purchasing pause, channel reallocation or a narrower assortment commitment.
Review price or exit
Repeated weak movement may justify examining price, placement, product role or exit options. Any markdown recommendation should consider margin and commercial constraints.
Document the finding, business implication, priority, recommended action and the condition that would change the recommendation. This keeps the output useful when new sales or stock information arrives.
Continue with the question that matters to you
Start with your decision
Apply the framework to your product data.
Elarva can review one focused question or conduct a deeper product and inventory audit. The available data determine the supported scope.