Comparability is one of the key concepts in BI Analytics for Retailers. If certain shops are new, closing down, being remodeled or closed due to whatever reason they cannot be compared the same way as the shops that are performing in a “business as usual” mode. In business planning and forecasting it makes sense to use history only for comparable stores. In other words, comparable stores can refer to a retail company's same-store sales compared to the previous year, and is used by analysts to make apples-to-apples comparisons from year to year.
Even though this sounds like a relatively easy task from a technical point of view architects may face a number of challenges when implementing a performing comparable store solution. For retailers that manage hundreds or thousands of stores it may be challenging to identify which stores are comparable at which point in time. Usually this relationship is maintained in a spreadsheet by analysts with a specific business knowledge. We may want to assign a comparability value by year to each store, such as: “Comparable”, “New”, “Closing down”, “Remodeled”, etc. Alternatively, in the case of a simplified approach Comparability attribute may be assigned “Yes” (comparable) or “No”. Comparability master data may be redefined and reuploaded several times during the year.
Below I review a few practical scenarios and architecture examples on how comparable master data can be used in analytics in a performing way within SAP BW on HANA (BW/4HANA) and/or SAP BPC Embedded.