Fidelity IT Leader Shares Hard-Learned Data Mapping Lessons

Loraine Lawson

Data mapping is hard, expensive and requires constant maintenance, but if you get it right, it can pay off in all sorts of interesting ways, particularly in managing data during mergers and acquisitions, says Bruce Phillips, the Information Security Manager at Fidelity National Financial.

 

In a recent interview with CSO Online, Phillips shares the details of the company's data mapping project, including their missteps in trying to build their own system.

 

Data mapping can be done for regulatory, legal or security reasons. But it's also first step in a range of data integration-related projects, such as transformation or migration of data, according to Wikipedia, which defines data mapping as "the process of creating data element mappings between two distinct data models."

 

As you might expect from that definition, it's hard to get right and it's expensive-particularly when you have to redo part of the project, as Fidelity National Financial did. So by sharing the company's mistakes and what they learned, Phillips is offering something pretty valuable here. It's the type of useful insight you seldom see with something as low-profile as data mapping, and I think business and IT executives will find it helpful.

 

While I suggest you read the full article, particularly if you're thinking about a data mapping endeavor, here's a highlight of some of the lessons Phillips shared from Fidelity's project:

  1. Don't do it unless you have multiple business reasons and multiple parties that would benefit. While data mapping can aid IT and help with legal or regulatory issues, Phillips says it's too expensive to undertake for just one reason. " If you don't have multiple constituents, don't try it. That's my advice to anyone. Unless you have a lot of uses for it, it's just too hard to do," he warns.
  2. Buy, don't build-unless development is your core business. Fidelity spent a year and a half trying to custom build solution. In the end, they had to admit that was a mistake and buy a tool. "Unless you are in the business of building applications, and we are not, it's too complex."
  3. Data mapping is an ongoing commitment-not a one-time project. "Once you create it, it's a living and breathing thing. You have to keep it up and maintain it. You must have a commitment to add resources and staff and time to just manage the data map itself, no matter what you are doing it for," warns Phillips.
  4. Know who's the steward of the data and then make that person responsible for updating the data model.
  5. Automate reminders for updating the data model as needed.


Add Comment      Leave a comment on this blog post

Post a comment

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

null
null

 

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.