Is there a single model of your enterprise?

The above question was posted by Donna Burbank in the CA Erwin Modeling group on LinkedIn last week. She has gotten quite a bit of responses. It is the hottest topic right now in this community. All of the comments are good reads and give good insight into the diversity of our community.
Some of my observations from the discussion-to-date:
  • Preaching to the choir. The majority of respondents definitely see the value of an enterprise logical model. If and how this model is manifested in an organization is driven by a variety of factors. Many corporations just are not structured to support such an effort. A common theme is “I would love to have that model but we do not have the time and resources to pull it off.”
  • Chicken or the Egg? What comes first? The approach to building that enterprise model seems to fall into two camps. Investing resources up front to build the model has long term paybacks that some organizations value. Many organizations have no interest in building a model through an infrastructure project and would rather build it piece by piece.
  • Does the business care? Conceptual modeling is a contentious issue. It is seen by many as a Visio like diagram that is expressed in business language and managed by or used primarily by the business. Others see it as a high level ERwin logical model to drive logical and physical data model development. Others see it as unnecessary and employ two model architecture of logical and physical models.
  • Theme park caricature or a Mona Lisa? The complexity and degree of detail in a conceptual enterprise model varies widely. Some see it as a blueprint for the design of enterprise data that only calls out major attributes. There are some that fully attribute this model and link it to project-based logical model entities and attributes. Many variations occur that fall somewhere in-between.
  • Is this relevant today? This is the typical reason offered up to “Why I don’t create a conceptual model today.” Agile development and faster turnaround are often cited as reasons to shortcut traditional project deliverables. Is the enterprise model one of the chopping block items? This is of course is highly dependent on the individual development environments. 
  • My hands are tied. Many companies are using ERP applications or industry models. Their deployment of an enterprise model is driven in these cases. Some chose to use the delivered model as their conceptual model. Others extend and interface with these models.
The above are just some of my observations by participating in this thread over the past week. Is there a single answer to Donna’s question? No. As a skilled data professional, we need to speak for the data. We are the voice that turns that data into information. We know data architecture and methods. The enterprise relies on our experience to make the correct choice for the enterprise.
Guiding data design with an enterprise point of view is a primary role we are given. How this role appears in your organization is largely in your hands. We need to look out for the company’s bottom line while using resources efficiently. Surf on over to this discussion topic and get a feel on how your fellow data community members are doing this in their organizations. This discussion is a perfect opportunity to see how our skill sets can be used differently to achieve the same result.
Tom Bilcze
Modeling Global User Community President
Find out more about “Why Be Normal?” at Want to know what “We Be Normal!” is all about? Visit to get the whole story.
Follow ERwin online through Twitter, Facebook, and LinkedIn.
Twitter: @ERwinModeling and #YBNormal
Print Friendly, PDF & Email

Leave a Reply

Your email address will not be published. Required fields are marked *