FAQs

> Contact Us

  • What is product data?
  • Product data is the formal name for material data or part data. In most cases it is comprised of free text descriptions which are found in the ERP or other IT systems. It includes all the information related directly to the nature of the product/material/part, such has manufacturer, sizes and materials. It does not include logistic data such as quantity, warehouse, price, etc.

  • What is data quality and what is product data quality?
  • Data quality is the tools and processes which support the creation of data according to well defined standards and regulations. Quality data should be complete, correct, accurate, consistent, conforming, clear and not redundant. Product data quality is a sub-domain of data quality related to product data.

  • What is taxonomy and why it is so important?
  • Taxonomy is a science of classification according to a pre-determined system. It is the framework and infrastructure of product data quality. Taxonomy consists of categories, categories' hierarchy, the attributes of each category and possible values for each attribute in each category. Good taxonomy should be simple to use and remember. It should also always lead to the same result when classifying or searching products. An inferior taxonomy will always lead to duplicate products, wrong decision making and excessive costs on logistics and operations.

  • How does product data become low quality at the first place?
  • There are many reasons for the low quality of product data in companies. Common examples are: lack of standards and regulations, lack of an enforcement system, lack of information, multiple languages and/or dialects, different naming conventions, limited space for describing complex technical products, reliance on acronyms and synonyms, different measure units (metric vs. inches) and different engineering standards (ISO vs. DIN).

  • What is the damage of low product data quality?
  • The simple and most apparent damage is higher sourcing costs and higher inventory costs. Other damages are higher operation cost and lower efficiency, lower customer service capabilities and faulty decision making.

    Low product quality means the inability to find the right product in the right time, hence, creating duplicate or similar products, building redundant stock, suffering from inefficient sourcing (buying same parts from many vendors with different prices), paying for unnecessary storage, logistics costs and many other cost factors.

  • What is product data cleansing?
  • Product data cleansing is the process of converting low quality product data into reliable data which is complete, correct, accurate, consistent, conforming, clear and not redundant. In order to do this correctly, the products should be classified by a proper taxonomy. Values need to be extracted from the original free text description, normalized and assigned to the right attribute. Products are then compared using their technical attributes and functionality in order to eliminate duplicates or similar products and produce new standard product descriptions.

  • Why is cleansing product data such a challenging task?
  • Cleansing product data is a very complex tasks because it involves extensive engineering knowledge and unreadable text in multilingual formats. Low quality data is created by human engineers and users using different languages, varying naming conventions, acronyms, synonyms while relying on different knowledge, experience, motivation and understanding. It is a real challenge, often described as "making an egg from an omelet!"

  • Why is manual data cleansing not a real solution?
  • To put it simply, we believe that the human being who created the mess in the first place is not capable of accurately fixing it. This human being is always subjective, basing judgment on his or her own knowledge and experience and therefore yielding constant inconsistent results. Manual cleansing can be done but only to a certain extent, that is not real cleansing.

  • What is the difference between product data, material data and part data?
  • No difference. They are all valid industry terms.