Editor's note: This column was contributed by Kim Stanick, a managing consultant with Baseline Consulting. She may be reached at kimstanick@baseline-consulting.com.

By the time a company decides to bring in data warehouse (DW) expertise, it may have lived with a problem for a long time and wants it solved as soon as possible. Three different clients recently shared this dilemma and approached us for advice. Notwithstanding the fact that their data warehouse technologies and designs were different, as were their data warehousing experience levels, they were all frustrated with their lack of progress.

Each client had reached a point where sustaining the existing production data warehouse environment was becoming unmanageable. Each faced a growing backlog with no systematic approach to address it. The business complained that it took too long for IT to deliver applications and that it always cost more than planned. "There's a saying around here," one client remarked. "Whatever the task, the estimate will be one year and a million dollars."

Each client shared hope that there was a better way to approach its data warehouse, and that with the right adjustments, a more manageable environment would allow them to meet the needs of the business more consistently. They had no time to research their data warehouse world as it could be.

Each situation turned out to be a good fit for our business intelligence (BI) scorecard service, which compares a client's data warehouse environment to best practices in six areas: requirements, data management, database platform, ETL (extract, transform and load), data delivery, and production management and support. The scorecard is a quick, focused service where we immerse ourselves in the client's environment to report findings and recommendations with a suggested course of action in just a few weeks.

The engagement is a flurry of activity, interviewing dozens of stakeholders and reviewing sometimes years' worth of artifacts. Most people are eager to tell us their stories.

As we dug in to each client's environment, some common issues surfaced. With deeper analysis, two overarching themes became strongly apparent: process and governance.

Each client complained about their inability to consistently deliver on time and on budget. Lack of process rigor included evidence such as:

  • The requirements process was loose or not followed, which led to extensive rework and delays; there was no accountability for process execution; and documentation handoff between groups was sloppy and inconsistent.
  • ETL was a "black box" - not well documented and not well understood, which made problem identification time-consuming and difficult.
  • Source system documentation, knowledge and relationship was poor, which led to significant downstream vulnerability for the DW team.
  • Data quality processes and oversight didn't exist or weren't consistently followed.

Each client also suffered because of the funding model. Development groups were given full responsibility (and blame) for prioritizing their work, and there was limited ability for business to scrutinize budgetary progress. Funding was for new development only, which caused a sort of catch-22:

  • Shared services for proper data management (data quality, data architecture, database administration, etc.) were not funded, so they didn't exist. This propagated end-to-end (stovepipe) development tactics, which hindered ability to achieve economies of scale.
  • Developers were also the resource for sustaining and remediation, which significantly impacted delivery time frames for new development.

These areas are quite significant. Solid processes are the foundation for DW success, and funding is the economic base that influences all data warehouse decisions.
The BI scorecard service concludes with a set of tailored recommendations. Not all of the areas that could be improved need immediate attention. It is important to establish a phased approach that considers culture, skills and technology. Hence, although their complaints were similar, our recommendations were tailored to each client.

For example, while one client lacked formal requirements processes, the other two had reasonable processes that they just weren't following. The recommendation for the latter included good old-fashioned management rigor - project management and accountability - but the former mandated requirements expertise to significantly bolster their process and oversee its execution.

The governance recommendations were customized as well. One client's executive steering committee had fizzled after a staffing turnover, so we recommended reinstating the charter. The other two clients didn't have a process or committee in place, and their business constituencies were very distracted. We suggested a BI portfolio approach to establish priorities. Other recommendations included hiring key roles for data management, reasonable padding of delivery estimates for support overhead and assigning a rotating sustainer.

When we describe the client's world in their own words, they can see we understand their situation. Translating their internal terminology into common industry terminology lets them realize that they are in a natural phase of data warehouse evolution. When we describe contributing factors and root causes, and offer solutions that have worked for other organizations that at one time shared their plight, they see the potential for positive change. Then they begin implementing the tactics. Problem solved! 

Kim Stanick is a managing consultant with Baseline Consulting. She may be reached at kimstanick@baseline-consulting.com.

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