Over the past few months, we have explored the concept of an enterprise data strategy or how companies plan to leverage and use data by turning that data into useful knowledge, thereby producing measurable improvements in business performance. There are several important stages to developing this strategy, including determining the analytic capabilities to be delivered and then deciding on the structure and management of the data. The action plan is the vital stage that makes the strategy work.
The action plan begins by determining the business problems that will be addressed by the enterprise data strategy and then organizing how the right people will drive and sustain the work. This month's column looks at the process, technology and delivery schedule.
New strategies in many companies are adopted to meet a new marketplace challenge. Typically, data strategies are created because the current corporate data must change or adapt to support the business in a new way. If the change is going to take hold and truly transform the business, new business processes must be defined. In other words, the company must work in new ways a new strategy with old processes is likely to fail.
For example, suppose the primary challenge for a company is the lack of standardization around customer information. The enterprise data strategy might then call for centralized assignment of customer identifiers across the corporation. Given that strategy, new processes would be needed to ensure that the numbers are assigned appropriately. Other examples of new processes might include:
- Avoiding entry of duplicate transactions.
- Using new query tools.
- Establishing data retention rules and archiving/retrieval processes.
- Maintaining entries in a common product catalog.
Companies must explicitly look at the enterprise data strategy and consider the new processes that will support it. This is a critical step to making the action plan a success.
Figure 1: Enterprise Data Strategy
At this point, when the relevant business challenges have been noted and the right people and processes are in place, the enterprise data strategy must examine the technology requirements. The action plan should consider a series of questions: What will the data architecture be? How will data be organized? How will it be delivered? How will it be moved or replicated or synchronized? How will it be accessed? (For more information about these questions, see my September 2001 column). The answers to these questions begin at a conceptual level, but must quickly be applied to the specific technologies as well. Sometimes the technology choices are defined as part of strategy development. More often, they are defined as future tasks in the action plan. The strategy cannot answer the technology questions completely. Therefore, the action plan must address the tasks needed to drive the technology decisions forward.
Common technology tasks are tool evaluations or proofs of concept for new technologies. Examples of product selection that may be named in the action plan are: meta data products, data warehouse DBMS (which may require benchmarking if the warehouse will be in the terabyte range), data mart DBMS, offline storage, user tools (application-building query environments, OLAP tools, ad hoc query tools, report writers) and ETL tools.
The final component of the action plan is the delivery schedule, which adds credibility to the action plan. Without the delivery schedule, efforts to move the strategy forward could be delayed or even cancelled. The longer the enterprise deliberates on the plan, the more it loses credibility and becomes out of date. It is vital that action begins quickly and that the plan delivers some "quick hits" to create value in the short term and demonstrate the viability of the strategy.
While a strategy offers a grand vision that inspires, a delivery schedule must show real progress in short intervals typically two to six months. Some of the action items will require more time to complete. However, it is important to keep the momentum going by delivering shorter tasks in parallel with longer tasks. Delivery of shorter tasks will continue to build credibility, while the longer tasks address major components of the plan.
One of the challenges in developing a workable enterprise data strategy is the fragmented nature of data in today's corporate environment. The answer to that problem lies in the fact that much of the data in an enterprise does not need treatment at the enterprise level. What data should be part of an enterprise strategy? More about that next month.
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
Already have an account? Log In
Don't have an account? Register for Free Unlimited Access