Success factors for clean product data

February 5 2020

Cleaning product data isn’t something that you can simply start and finish between lunch and afternoon tea. Ensuring optimal data quality is serious work that takes time. However, it’s well worth doing, as it provides substantial benefits: improving findability of products in webshops, improving tender offers, increasing margins, and providing better insight into company inventory.

At Mydatafactory, we know the advantages of clean product data like no other, and we understand the factors for success in data cleansing due to our vast experience with customers. Hard-won wisdom that can only come from time spent in the trenches, as some might say. Below, we share the first three success factors with you. Do you know what makes or breaks your product data quality?

1. Human action

The quality of your product data is largely determined by the way people input, supplement, or value product information. About 98 percent of the problems with data quality are due to the people working with product data. Without human action (more specifically, their mistakes), hardly any data quality issues would occur. Fortunately, it’s also through people that mistakes or gaps in data quality can be fixed. Human action is absolutely a factor for success with regards to data quality.

For Example: Suppose someone types “bule” instead of “blue” in the color attribute entry for a product. Your product data soon becomes contaminated on all your channels. Fortunately, this can be prevented by creating synonym lists in advance with common typing and spelling errors, which allows the software to recognize the errors and change them wherever they occur.

2. Identify who your product experts are and involve them in the process

Optimal product information is the foundation for your data quality. Clean product data is possible only when all information about a product is correct. Identifying who the real product experts are within your organization and involving them in the process so they’re able to give knowledgeable input therefore becomes a crucial part in cleaning up product data. It is these product experts who know which attributes are important for a certain item, in which product class the item best belongs, and the possible values the attributes should have.

For example: For example: In your organization, someone without specific product knowledge likely won’t recognize errors in product specifications. For him or her, an “A3 screw” might seem like a logical thing, because A3-sized paper exists. Only an actual product expert can check information for any mistakes and provide correct information.

3. Classification structure

Most wholesalers have an ever-increasing product range that they sell both online and offline. Our large international customers have a product range of more than 2.5 million items, so a good classification structure is essential. Without standards in your products and classification structures, managing product data and offering information in an unambiguous way becomes extremely difficult. A classification structure helps you organize products so that existing customers and potential customers alike can find the products they need in simple, understandable ways. Your customers’ experiences can be greatly impacted by the choices you make. Which classification structure (ETIM, GS1, eClass, EZ-base, etc.) would make the most sense and best fit your products, customers and internal processes? Which would be helpful in increasing the quality of your product data?

For example: When a customer searches for shower faucets, having the entire range of shower faucets grouped together allows the customer to more easily make a choice, as all the options are readily visible. Plus, when your company expands into other global markets, having international classification structures can help you communicate your products in different countries. Even if the information is in different languages and the products have different names (such as “lagers” in Dutch and “bearings” in English), the products would belong to the same class or group since you’re still talking about the same thing..

Want to know about the other success factors for clean product data?

Don’t want to miss knowing about the other four success factors for clean product data? Sign up for our e-mail alerts! Send an email to info@mydatafactory.com and indicate that you would like to receive updates from us.