Retail assortment planning software: what to evaluate

Retail assortment planning software should pass a test on your own SKU-store export before you compare presentation quality or feature lists. The right approach depends on whether your team needs to understand product performance, maintain an approved range, or execute purchasing decisions.

Start by writing the decision the buyer must make and the evidence needed to approve it. The separate retail assortment planning workflow explains how to prepare that decision; this page focuses on evaluating the software supporting it.

Compare approaches by the work they must complete

Do not treat every product with a retail dashboard as an assortment system. Group the options by the job you will test, then ask each supplier to demonstrate the relevant behavior. The table describes evaluation requirements rather than guarantees about any particular product.

One team may need more than one approach. A reporting tool can help explain why an item needs review while the purchasing system remains the place that issues the order. Use the broader sales-planning software guide for general workflow selection, then apply merchandise-specific checks here.

Keep responsibilities explicit during selection. Merchandising owns range coverage, finance owns the spend limit, and operations owns the receiving and store-execution checks. A supplier should demonstrate the handoff between those owners rather than leaving it as an assumption.

Prepare a small export with known answers

Use one period and retain the product and store identifiers, including leading zeros. GS1 US explains that a GTIN identifies a product across retailers while an SKU is an internal identifier. Your evaluation should prove that the software keeps the identifiers needed to join your files without silently merging variants.

A fictional US home-goods retailer uses this four-week baseline. Available days describe when sellable stock was available, not simply when the store was open. These are test inputs supplied by the retailer.

Specify net sales after returns and discounts. The IRS merchandise gross-profit calculation subtracts COGS from net receipts; the evaluation should preserve that distinction rather than subtracting purchases made during the period. Use the company's own consistent COGS policy.

From those rows, the expected results are:

Gross profit is 4,000 − 2,440 = $1,560. Company margin is 1,560 / 4,000 = 39.0%. The simple average of the row margins is 38.3%, rounded to one decimal place, and is the wrong company result because the rows have different sales weights. The gross margin guide owns the full calculation and reconciliation.

Weeks of cover divide closing units by units per available week. North A is 35 / (70 / 28 × 7) = 2.0 weeks. South A has zero closing stock, but its observed weekly rate comes from only seven available days. A useful evaluation preserves that caution when displaying or explaining the rate.

Run exception tests before scoring convenience

Ask each supplier to load the same baseline, produce the expected figures and expose the underlying rows. Then make controlled changes to the file. Record the result and the person who observed it rather than relying on a sales presentation.

1. Repeat a row. Add an exact duplicate of North A with the same source record key. Counting it twice raises sales to $6,100 and gross profit to $2,400. Require a warning or an explicit duplicate-handling step; automatic deletion also needs a clear rule because identical amounts can represent legitimate separate records.
2. Remove an availability value. A missing value must remain visible as unknown. It must not silently become a full selling period or zero demand.
3. Filter to North. The result should contain only the two North rows. A spreadsheet baseline can verify sales with SUMIFS; its official function documentation specifies conditional sums across criteria.
4. Change a buying constraint. Supply a purchase budget and a supplier carton multiple. Ask the supplier to show the proposed quantity before and after the change, its cost and any exception. Test the actual constraints your buyer uses rather than accepting a generic optimization claim.

For the North filter, put store in column A and net sales in column D, with baseline rows 2 through 4:

=SUMIFS(D2:D4,A2:A4,"North")

That returns 2,100 + 900 = $3,000. Covirage is not affiliated with Google.

The inventory assortment guide covers the underlying breadth and depth decisions. Software should preserve the buyer's assumptions and approval evidence when that discussion changes a decision.

Separate required capabilities from preferences

Mark a capability as required when its absence prevents the team from approving a range safely. Examples include preserving store-level detail, reconciling dollars to the export and identifying missing availability. Treat display preferences separately so an attractive chart cannot compensate for an incorrect total.

For each requirement, record pass, fail or unproven, plus the file and screen or export that supplies the evidence. A claim remains unproven until tested. Review access controls, handling of uploaded files and exportability against your organization's procurement requirements using a sample of your own data that you are authorized to share.

Ask for the complete cost proposal and the work needed to maintain the process. Include training, product hierarchy changes and the recurring effort to prepare files. The selection should fit the buyer's working calendar as well as the feature checklist.

Where evaluations go wrong

• Category totals hide variant errors. Matching total sales does not prove that sizes, colors or stores were assigned correctly. Trace individual rows too.
• Stockout sales become a demand promise. South A's observed rate needs context. It does not establish what a fully stocked future period will sell.
• Analysis is confused with execution. Require evidence for order creation, approvals and operational handoffs whenever you need those functions.
• A clean baseline is the only test. Duplicate records and missing values reveal how the software handles the files your team actually receives.
• The supplier sets every assumption. Use your own product hierarchy, constraints and expected figures so the comparison answers your requirements.

Analyze your own assortment export in Covirage

Bring a sample of your own SKU-store performance export to Covirage for the analysis part of the review. Covirage maps the columns, its tools compute and check the figures, and the external AI model explains them with every figure traced to its rows. Create a free Covirage account, which takes about a minute, then set up your business and bring your file.

See self-service analytics for the supporting approach to reviewing your own business files.