Retail comparable sales: rebuild the store cohort
Comparable retail sales measure change within a defined population, not all growth in a retail chain. Rebuild the population and calendar before interpreting the number as stronger customer demand. Store openings, online fulfillment and price changes can each alter the headline without answering the same operating question.
Fix the cohort and sales boundary
Walmart’s annual report explains that comparable sales include e-commerce and discusses fiscal-calendar versus retail-calendar comparisons. This company example shows why an “online contribution” should not automatically be added to a comparable-sales number that already includes online sales. Other retailers can define eligibility differently.
Record the minimum store age, closure policy, remodeling treatment, fuel inclusion and currency scope. Specify whether relocated stores keep their comparable status. A cohort excluding every difficult renovation may describe continuing locations well while omitting an important cost of maintaining the network.
Hypothetical transactions and ticket
Assume the same 80 mature stores generate one million transactions at an average ticket of 20 in the comparison period. In the next equal-length period, transactions fall 2% to 980,000 while average ticket rises 5% to 21. Sales increase from 20 million to 20.58 million, or 2.9%.
The exact growth relation is (1 + transaction growth) × (1 + ticket growth) − 1. Here, 0.98 × 1.05 − 1 = 2.9%, rather than simply adding −2% and 5%. A positive comparable-sales result can therefore coexist with fewer purchases.
| Cohort worksheet | Prior period | Current period | Interpretation |
|---|---|---|---|
| Eligible stores | 80 | 80 | Same population |
| Transactions | 1,000,000 | 980,000 | Down 2% |
| Average ticket | 20 | 21 | Up 5% |
| Comparable sales | 20,000,000 | 20,580,000 | Up 2.9% |
| Sales from new stores | Excluded | 2,000,000 | Separate network expansion |
Suppose 3 million of current cohort sales are fulfilled through stores after online orders. If that amount is already included in 20.58 million, adding it again would overstate sales. The appropriate question is channel attribution within the same total, not the sum of overlapping headlines.
Check calendar and product effects
A 53rd week, holiday shift or different trading-day count can affect the comparison. Use the issuer’s same-calendar series where available, and explain any reconstruction. Matching quarter names is insufficient when the actual selling weeks differ.
Average ticket is also not pure pricing. A household buying fewer but more expensive items can increase ticket while unit demand falls. Add items per basket, category mix and disclosed price effects where evidence exists. If these components are unavailable, state that the ticket increase cannot be divided precisely between inflation and mix.
Make the demand question specific
If the mechanism is repeat customer growth, comparable sales alone may be insufficient: transactions can include repeat purchases by fewer households. If the mechanism is mature-store productivity, sales per trading day and selling area may be more relevant. Choose the observation that matches the claim rather than treating every growth measure as interchangeable.
The next review should preserve the same store-age threshold and channel treatment. A changed cohort can warrant a bridge or a restarted series. This worksheet evaluates retail activity; it neither forecasts consumer spending nor establishes that a positive comparable-sales figure produces higher profit.
FAQ
Should new-store sales be added to comparable growth?
Use a separate company-sales bridge. Comparable growth is a percentage for a specific population, not an amount that can be added directly to new-store revenue.
Does higher average ticket prove pricing power?
No. Product mix, basket size and discounts also affect ticket. Inspect unit and margin evidence before assigning a pricing explanation.