No matter how good a Master Data Management matching process that you have, there needs to be some commonality in the data for records to match. If two records have different company names, addresses, phone numbers, website addresses, or other information that may be used for matching purposes, then there is not any way that your Master Data Management match project can bring these two records together based on an automated match.
For example, DePuy Orthopaedics, located in Warsaw, IN, is a subsidiary of Johnson & Johnson. Ethicon, located in Somerville, NJ, is also subsidiary of Johnson & Johnson. However, if you do not have any information in your databases indicating that these organizations are related, it is going to be impossible to put them together with your Master Data Management match. Consequently, if someone wants to know how much business the company is doing with Johnson & Johnson, they are not going to get a complete answer.
If you want to put corporate hierarchies together with subsidiaries that do not share any common information, then you are going to need an outside source. An external marketing file that contains corporate hierarchy information can be a great resource. However, sometimes, you only need to create hierarchies for a relatively small sample set. Such as, if you want to answer the question, “Who are our top 20 customers?”, you do not necessarily want to purchase an external list just for this purpose. Instead, you may want to research the top 20 – 50 companies from your Master Data Management automatic match and then augment the results with manual research.
Of course, a company’s website is a great place to start. However, sometimes a company website does not layout the corporate hierarchies in a manner that is easy to research. Therefore, we have found that one of the best places to research corporate hierarchies is Wikipedia. Many times Wikipedia can lay out corporate hierarchies in a fashion that is quicker and easier than a company website. Sometimes, it is more complete. Also, sometimes a Wikipedia entry will provide history. So if your data includes a record for a company whose name has been changed after an acquisition, sometimes Wikipedia may have this information where the company website may no longer include this. Of course, Wikipedia is not verified, so it’s important that you double check any findings on Wikipedia before incorporating it into your Master Data Management match.
I am mapping the military/industrial complex based on higher hierarchical structures that are indeed very difficult to weed out because 'they' obfuscate the links and relationships in a complex web of seemingly unstructured partnerships. Boeing is one of them. How would I tap into 'data silos' any differently by using your software?
Posted by: Mark Golding | May 18, 2009 at 11:34 AM