FAQs - PDI

Are large chains such as 7-Eleven, Buc’ees, Pilot etc. included in these data?

The data that PDI is providing is primarily from the category of independently owned convenience stores, which is the largest segment of the c-store market in the US. The data for some of the large chains mentioned – 7-Eleven, Buc’ees, Pilot etc. are not available due to data rights restrictions – so they are therefore not in the data.

How is ShopperID determined?

  • ShopperID: Created by hashing the concatenated values of first 6 digits of the card (BIN), the last 4 digits of the card, and the state abbreviation of the store where a purchase was made.
  • Cash and EBT payments are not included in the models.

Why are cash payments with associated ShopperIDs?

  • Possible that this was the result of split payment where multiple payment types were used.

How are lottery items handled?

  • PDI notes that these lottery items appear to be seasonal, and so with out continuous GTIN mapping it is reasonable to see a decline in the number to GTINs in that library.
  • The last update to the library was the beginning of the year (2025), the decline is expected.

How can I filter lottery transactions that are mislabeled?

  • PDI confirms that filtering lottery transactions to the whole dollar amount can remove miscategorized GTINs.
  • PDI also suggests to filter those transactions associated with a discount to remove miscategorized GTINs.