DM Review Published in DM Review in October 2001.
Printed from

Meta Data & Data Administration: Repository: Buy Versus Build

by David Marco

More companies than ever before are reducing their information technology (IT) costs and delivering a competitive business advantage by building successful meta data repositories. The companies that are building these repositories are faced with a fundamental question: Should we buy a meta data integration tool and use that tool to build our repository or should we build a custom meta data repository? This article will address the four essential questions that will guide your decision. Before you can answer these four questions, there are some tasks that need to be completed.

First, define the specific business and technical requirements of your meta data repository. This is best done by completing a meta data repository project scope document. Do not look at tools until you have completed the project scope document. Second, an important aspect of the project scope document is that it identifies the sources of meta data you will need to integrate into your repository. Note that not all meta data needs to be stored in the central meta data repository. Third, decide how frequently you will extract the meta data from these sources. The frequency with which you will update the meta data within your repository is based on your company's specific business requirements for the repository. Fourth, gauge your corporate climate. Some companies always custom build their applications. Other companies rely on tools and prepackaged applications. Your company's climate could dictate your choice.

When talking about purchasing a meta data integration tool, it is important to understand what you are buying. First, all of these tools come with their own meta models. A meta model is a fancy term for a physical data model that holds meta data. Second, these tools typically come with prebuilt impact analyses that pull meta data from the base meta model. An impact analysis is a technical meta data-driven report that aids a company's IT department in drastically reducing the time for IT development. (For examples of these reports, see my September 2000 DM Review column, "Meta Data ROI: Making Your IT Department Better, Stronger, Faster.") Third, a meta data integration tool has the ability to bring in specific sources of meta data and integrate them into a meta data repository. In addition, as these sources change over time, the meta data tool vendors will enhance their tools to accommodate changes. Do not underestimate the task of keeping your repository's integration processes up to date with the changes that will occur to your meta data sources.

Question 1: Can your company afford $500k $750k for software? The cost of purchasing a high- end meta data integration tool typically starts at $500,000 and can go as high as $850,000. In addition to this purchase price, the annual maintenance fees can run anywhere from 15 percent to 25 percent of the original price of the tool. If your company cannot afford this type of purchase or will not allocate this amount for software, then you should consider a custom-build or a lower- cost meta data integration tool.

Question 2: Can you assign dedicated resources to become knowledgeable about the tool? All meta data integration tools require that multiple dedicated resources be assigned to work on the tool. If you do not have resources to allocate and do not have funds to get these resources from a vendor (software or consulting), then you should not be purchasing a tool. Like any other major IT initiative, meta data repository projects require dedicated resources. If you do not have dedicated resources for your project, then you need to have a very limited project scope.

Question 3: Is there a tool that certifies enough of your key meta data sources? When meta data integration tool vendors state that their tools certify a specific source of meta data, it means that the tools have an existing process that pulls meta data from this source and loads it into the meta model. (For more on this topic, see my December 2000 column, "Evaluating Meta Data Tools.") It typically does not require a great deal of effort to load certified sources of meta data into the repository.

Question 4: Will your company save enough development costs (current and future) to justify the tool's purchase? If the tool does not certify a sufficient number of meta data sources to justify the cost of the tool, you will most likely want to do a custom build; however, if you are starting your project small and want to grow, a tool may be a good choice.

If you have answered no to any of these four questions, you should not be purchasing a meta data integration tool. On the other hand, if you've answered yes to all of these questions, a tool is most likely in your future. As a guideline, at Enterprise Warehousing Solutions, Inc., approximately 60 percent of our meta data repository clients select building over buying. Keep in mind that most of our clients tend to be conducting larger meta data repository initiatives.

David Marco is an internationally recognized expert in the fields of enterprise architecture, data warehousing and business intelligence and is the world's foremost authority on meta data. He is the author of Universal Meta Data Models (Wiley, 2004) and Building and Managing the Meta Data Repository: A Full Life-Cycle Guide (Wiley, 2000). Marco has taught at the University of Chicago and DePaul University, and in 2004 he was selected to the prestigious Crain's Chicago Business "Top 40 Under 40." He is the founder and president of Enterprise Warehousing Solutions, Inc., a GSA schedule and Chicago-headquartered strategic partner and systems integrator dedicated to providing companies and large government agencies with best-in-class business intelligence solutions using data warehousing and meta data repository technologies. He may be reached at (866) EWS-1100 or via e-mail at

Copyright 2004, Thomson Media and DM Review.