Store openings and closures: keeping a panel comparable

A panel that changes its members every month can't tell you much about growth. In retail data, a panel is the set of stores whose sales are being measured. Store panel changes, such as new stores joining, stores closing or stores dropping out of the data temporarily, move totals up or down without any change in how existing stores are performing.

Keeping a panel comparable is one of the least glamorous parts of retail analysis. It's also one of the most important.

What is a store panel?

It's the list of stores included in a dataset for a given period. A panel might cover all stores that reported data that month, or a fixed set chosen for a study. The choice affects every total built from it.

In a network of independent retailers, the panel shifts constantly. New stores come on. Some close. Some change owners, remodel or switch systems. Some have a week where their data doesn't arrive. Each change alters the total.

Suppose a panel has 100 stores in March, each selling an average of $10,000 in a category. That's $1 million. In April, 110 stores report, each selling the same $10,000. The total is $1.1 million. Nothing changed in any store. The panel grew.

How do openings and closures distort results?

Openings inflate totals. New stores add sales that didn't exist in the base period. If you compare this month's total with last year's, some of the apparent growth is simply more stores.

Closures deflate them. Stores that closed contributed sales last year and none this year. Their absence looks like decline.

New stores also behave differently. A store in its first months may still be building its customer base, stocking up or adjusting its assortment. Its sales don't reflect a mature store. Mixing new and established stores in the same average blurs the picture of both.

How does a same-store comparison fix this?

A same-store sales comparison includes only stores that were open and reporting in both periods being compared. New stores and closed stores are set aside. What's left is a like-for-like view of how existing locations performed.

The NRS Insights monthly report is built on this principle, which is why it's called a same-store sales report. You can see the approach in the July 2026 same-store sales report.

There are choices within the method. How long must a store be open to count? Many analysts require a full year, so that the store has a comparable base period. What about a store that missed a few days of data? Some methods drop it for that period, others fill small gaps. The rule should be written down and applied consistently.

What else can break panel comparability?

Temporary gaps in reporting. If a store's data doesn't arrive for a week, that store should be handled consistently, either removed from both periods or flagged. Silently including it with zero sales creates a false decline.

Remodels and ownership changes. A store that closed for three weeks to remodel isn't comparable during those weeks. Some methods exclude it until it has a full comparable period again.

Changes in what's measured. If the panel's stores start carrying a new category, category totals rise even if nothing else changed. That's an assortment change, not a same-store gain in existing products.

Handled carefully, panel rules make trends trustworthy. Handled loosely, they make every chart suspect. The NRS Insights reports describe comparisons on a same-store basis for exactly this reason.