LESSONS - Take a Different Approach for Near and Long Term Data Warehouse Success
By DATAllegro, Inc.
Data warehousing is a mature industry with hundreds of vendors and thousands of consultants. There are plenty of well-thought-out data warehouse (DW) implementation methodologies developed by people who have “been there, done that,” and done it well. Best practice awards can be found on nearly every platform and software combination and across all data volumes. And most people agree on the basics of due diligence, design, and development tactics. With all of this expertise and knowledge, still, there are bumps along the road. Why?
In light of all of the product and methodology enhancements over the last 25 years, there’s one element of data warehousing that hasn’t changed, and that has the most significant impact on project outcome: The human element. I’m not talking about knowledge; I’m talking about character. The people tasked with bringing forth the DW are the most critical success factor, yet they are not typically thought of as such when companies are considering their data warehouse approach. I’m referring not only to the members of the technical DW team, but also to all of the people across all of the departments who will be tasked with some form of DW ownership, responsibility, or commitment.
Most companies focus on the technology. Instead, focus on the team. Think of building a DW as if it’s a lifelong relationship (it is). DW teams, like any relationship, need nurturing. Most pitfalls can be attributed to differing perspectives or poor communication. Occasionally, it’s something deeper.
What are some of the human issues and how are they overcome?
Overcommunicate. Sometimes we think we’re speaking the same language, only to discover we’re not. Better to risk redundancy than to leave holes.
Place overall responsibility and adequate authority with one person. Provide undisputed clarity of roles and ownership with no gaps and minimal overlap. Establish agreed-upon processes. Don’t shoot the messenger who brings bad news.
Air all concerns and get universal buy-in frequently. Establish goals that reinforce straightforward behavior.
Management must incent all parties connected to the DW such that they pull together, not apart, and place the same priority on the DW effort.
Poor team dynamics
Communication helps here, but sometimes there is a strong personality misfit or even sabotage. This must be corrected. A third-party consultant can help, but don’t expect them to work miracles around a severely dysfunctional team.
Most companies focus on the technology.Instead, focus on the team. Think of buildinga DW as if it’s a lifelong relationship (it is).
What are additional positive actions you can take to help the DW team be successful?
- Form a tech-business “bridge” organization to span the gap and ensure good communication between business and IT. A strong, business-savvy person who has jumped over to IT or vice versa is a star player here.
- Tie bonuses, including those of top-level management, to the success of the DW. Incent all of the constituencies so that they have the same goals.
- Eliminate unhealthy bias in the technical evaluation. To get your business, each vendor should be willing to perform a proof of concept. Give each vendor identical requirements and time constraints and stick to them.
- Reduce the number of technology components to oversee (fewer moving parts). Preintegrated technology from a single vendor can help by providing “one throat to choke.”
The human element is one big reason why enterprise data warehousing (EDW) is so difficult. Getting people to agree across the enterprise, such that a data model, physical schema, and metadata can be created, is hard. Unfortunately, there are no DW therapy books. Remember that a stellar team can overcome mediocre technology, but even great technology cannot hide a dysfunctional team. Pick stars.
This article originally appeared in the issue of .