Continue in 2 seconds

Building Business Intelligence

Published
  • October 01 2003, 1:00am EDT
More in

Working with an advisor can pay tremendous dividends, but often it requires a step back from the day-to-day to work the advisor into the project for the long-term benefits. A little preparation can go a long way. Usually, there is advisement not only to the project/program, but also in how to work with an advisor.

By advisement, I mean having an outside specialist observe, challenge and teach throughout the life cycle, from requirements through the early stages of production. The advisor could also fill gaps as required in producing deliverables for the project. Advisement is distinguished from other forms of consultancy such as requirements gathering or other early-stage steps in a project that focus resources on a specific delivery. Although it may begin with an assessment of the program, it is distinguished from assessment or other point-in-time engagements by its ongoing nature.

Programs under quality advisement tend to avoid the usual slew of failure points for data warehousing – overscoped projects, incorrectly sponsored projects, failure to set and adhere to guiding principles, focus on technology – not business ROI, non-scalable architectures, lack of business involvement and inadequate team experience in delivering projects of this nature.

An advisor can assist with many aspects of a data warehouse program. Commonly treated as afterthoughts, these areas are necessary components of a program. These include guiding principles, readiness assessments, architecture specifications, meta data strategies, user management processes, data quality processes, security processes, data source analysis, conceptual modeling, testing procedures, customer boarding and training specification, customer support model specification, maintenance planning and support readiness assessment.

Programs with quality staff but lacking real-world data warehouse experience are well-served to rent the experience by utilizing an advisor. Large studies have found lack of experience to be one of the top reasons for data warehouse failure. Advisement helps mitigate this risk by seeding experience onto the team.

Furthermore, quality advisement will review all interim deliverables throughout the project. It is paramount to a successful advisement that it is not meant to just deliver paper. The build team should be prepared to react to feedback and iterate the deliverables until the concerns are met. The advisor should review the project plan, the organization, the user requirements, the data model, the source-target map, the data quality plan, technology planning and purchase orders, applications and reporting/user interface specifications.

Advisors must be willing to provide expected and unexpected advisement. Advisors who only parrot what the client is expecting to hear and do not provide unique insight into the program are not very valuable. Far too many business intelligence projects go awry because the build team is unable to deliver to expectations and the tough messages on the need for proper scoping, prioritization and business involvement. When these items are bypassed, it always comes back to haunt the project.

Clearly, an advisor needs to be flexible and well-rounded to provide advice in all areas, including technology, process, people and business issues. These components will come together to form a cohesive whole and should be considered as such by the advisor. "Advisement by committee" is not recommended. Having end-to-end experience at the project or technical lead or advisory level with numerous, similar projects is paramount to advisory success.

An advisor also needs to be able to step back and allow the client to perform all manner of the work possible. Advisors should be somewhat flexible with their methodology – understanding the battles worth fighting and those not worth fighting – keeping the end goal in mind.

An advisor can be more objective than an internal or consulting group manager and, while sensitive to the political situation of the client, can be more dispassionate and provide the right advice around tradeoffs to make for a fast and lasting return on investment. By knowing the program intimately, the advisor is ready to respond to any program question without having to start at square one. An advisor can also help transfer internal program responsibility when required.

Focused BI consultancies tend to have the people who can provide the best advisement by these criteria. For organizations that are not prone to outsource entire projects, advisement is a perfect opportunity to get the outside advice that remains loyal to the approach of building systems in house. Advisors also can be involved in programs that are completely or largely assigned to "big" consultancies. Advisors become the client advocate with oversight authority over the consultancy.

A successful engagement includes the advisor in weekly team calls and on all team messages and deliverables. Separate, private semiweekly discussions with the project leader also create opportunities to surface issues that are immediate or on the horizon.

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