A great way to sharpen our analysis and modeling skills is to continuously address real-world scenarios. A modeling scenario along with suggested solutions will appear each month in this Design Challenge column. I have been conducting this exercise on my Web site and thought I'd share the results with you.

The scenario is emailed to more than 1,000 designers up to the challenge. All responses, including my own, are consolidated, and the best will appear here. If you would like to become a design challenger and have the opportunity to submit modeling solutions, please add your email address at www.stevehoberman.com/designchallenge.htm. If you have a challenge you would like our group to tackle, please email me a description of the scenario at me@stevehoberman.com.

The design challenge presented in this inaugural column is from a recent consulting assignment I had with a large telecommunications company as it embarked on an enterprise data modeling effort. Building an enterprise data model (EDM) of an entire corporation is no small feat, and there can be a large initial investment in time and money followed by a maintenance effort for the life of the EDM.

The Challenge
You are the manager of this newly formed enterprise modeling group, and you need to "sell" the EDM to senior management. You plan on explaining the benefits of increased business understanding across departments, higher quality data and lower software development costs over the long term. But how would you answer the question, "What is the EDM ROI?"


There are two ways to solve this challenge:

  1. Find some way to calculate ROI.
  2. Refine their question.

Calculate EDM ROI

Responding to this challenge, Jan Kamil (senior data modeling consultant) and Yvonne Balditt (data analyst) proposed a detailed calculation of EDM which I found extremely valuable. Assume EDM ROI = (EDM contribution to profit) - (cost of the EDM). The contribution to profit can include factors such as the cost savings of staff time not spent on research because an EDM exists and a reduction of outside marketing costs because the data to support customer trends, such as retention and attrition, will eventually exist in the EDM's holistic customer view. For example, you can save $15,000/year paid to an outside marketing company for information that now exists within the EDM. The cost of the EDM asset includes not only staff wages, hardware and software, but also lost opportunity costs (the cost of the ROI of projects that staff is not working on because they're building the EDM).

Refine Their Question

You might agree with industry expert John Ladley's analogy, "ROI on a model is like asking for the ROI of the blueprint for a new factory being built. The ROI is on the factory, not the blueprint." Instead of calculating ROI, we can rephrase the question into, "You're telling me all these great benefits of an EDM. Show me some real proof!" Instead of trying to find success stories with an enterprise model (because there aren't many on the Web), use Google, DMReview.com and tdan.com to find case studies that highlight the value of high-quality data. If you can convince management of the value of high-quality data, you are a big step closer to convincing management of the EDM's promise for a single version of the truth.

I like to use the Hershey Foods turnaround story with implementing SAP as a great example of the value of data quality. Hershey succeeded in a large part due to putting their energies behind their maxim, "data is king."

Håkan Edvinsson is a data management trainer and consultant who addressed the EDM ROI question by showing how an improved enterprise architecture would solve management's most pressing problem for a telecommunications company. "We came with convincing arguments and solutions and explained what they felt were real problems. And we didn't scare them with cardinality."

Ben Ettlinger is a lead data administrator who illustrates the importance of data quality and, therefore, the enterprise data model with the following story. "NASA lost a $125 million Mars orbiter because one engineering team used metric units while another used English units for a key spacecraft operation. A good enterprise data model is not just a picture with boxes and lines; it includes metadata - detailed documentation about your data that describes what is contained in each data structure, including the unit of measure. A good enterprise model makes use of all the functions and features available to properly present the metadata of an enterprise. Give me a list of the mistakes your company has made, and I could tell you the ROI. NASA has a $125 million one." 

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access