Liliendahl on Data Quality

When working with data quality in the product master data management domain you are very dependent on your business partners. Product master data are shared along with the physical products in the ecosystem of manufacturers, distributors, retailers and end users.

In a current role, I have worked a lot with sourcing product data from suppliers. One of our recurring examples is about one of our product categories being toilet seats. In that context, we have three different kind of suppliers:

• Those who use the term “toilet seat” in their product descriptions. That is marvelous, then we can use that part of the product description directly as it is. Wonderful data quality.
• Those who only use the term “seat” in their product description. Well, it is not really bad data quality for a dedicated manufacturer of bathroom stuff, because what could a seat else be in that context. However, for consistency reasons we have to correct “seat” into “toilet seat”.
• Those who use the term “WC seat”. Actually, “WC seat” could be more accurate than “toilet seat”, because we are talking about seats for a room with water opposite to older solutions. Nevertheless, for consistency reasons we have to correct “WC seat” into “toilet seat”.

Manufacturers, distributors and retailers have to work together in order to create win-win situations by sharing product data with an optimal data quality. This is however not straight forward, as you always will be part of an ecosystem where your competitors operate too and often you are not prepared to share the same seat as your competitor.