Product feed layout overview

Overview

This document provides additional details related to the product feed layout for the retail industry, including the use of the feed layout, how the data is reflected in the database, and how to use the data in Redpoint CDP.

This section provides the following detailed information:

Feed layout intent

The intent of this feed layout is to allow for product/item details to be loaded to the Redpoint CDP. Product data is used to link the transaction detail records to the specific product information related to a purchase. This data is used to calculate aggregates related to an individual's purchases of specific products as well as within a given category. This allows you to target customers based on previous shopping behavior using this data to make product recommendations.

Extension table

This feed has an associated extension table.

Refer to extension tables for general information about using extension tables.

Feed layout areas of consideration

Consider the following when providing data for this feed layout:

• If there are multiple brands, then product will store data at brand level.

• In addition to active products, discontinued products or products no longer being sold should also be loaded to this table to ensure that when historical transaction data loads are performed, they will pass the referential integrity required by the transaction data loads. If the products are not identified when loading the transaction data, the records will be placed in holding tables.

• Discontinued products themselves may also be of interest for analytic or research related to products, even if they are not tied to historical transactions.

• In order for the product category summary table calculations to be made, there are two fields required to be populated in the product feed to support this:

  • product_category_code

  • product_category_desc

• In cases where all information in a new record other than source_rec_create_date and source_rec_update_date matches an existing record in the database, the database record will not be updated, and the new record will be discarded as a duplicate. This is to ensure that product based portions of aggregations are only regenerated when the actual product data changes. This means that the data in the CDP database will match the oldest record if the files are processed sequentially and the newest record if they are processed as a batch.

Identity matching

Product data do not go through matching.

Use in RPI

The product data that is provided can be used to create various selection rules in RPI. There are two primary summary tables that utilize the product information:

• Transaction detail summary: provides a denormalized view of the transaction data.

• Product category summary: provides aggregates at the product code level for individuals.

Use in aggregates

Summaries where the location is used.

• INDIVIDUAL_BU_PRODUCT_CATEGORY_SUMMARY

• TRANSACTION_DETAIL_SUMMARY
  A lot of the fields that are in the product table are found in this summary, so the more fields that you provide, the more complete the product information will be in this summary.