-
-
Marketplace
-
Channel Resources
Articles from this Site
-
One brand, one Web site! DM Review is now the home of all the content you're used to at BIReview.com and much more. If you are registered at BIReview.com, you're already registered at DM Review. If not, take just a moment to sign up for all the free services we have for you at the new DMReview.com.
Set Business Requirements First
Many organizations struggle to collectively define and articulate their business requirements for performance measurement and management. These business requirements are the basis for conveying key performance indicators (KPIs) and supporting information for monitoring, analysis, reporting, and decision-making purposes through business intelligence (BI) technologies. Some avoid formally defining or documenting their business requirements altogether, chalking the effort up as too costly or difficult. These challenges can stem from a lack of executive sponsorship or availability of stakeholders, broad scope, lack of business imperative alignment, too many performance measurements, too many systems to analyze, scarce expertise, competing initiatives, countless technologies, or endless debates over data modeling approaches.
However, Fortune 1000 companies increasingly need information for decision-making purposes to enhance their performance and reduce risk. Without knowing the changing information priorities of business executives and users, shifting market trends and new opportunities may not be detected - translating into a loss of action that could adversely impact performance. In an era when sustainable advantage is linked with an ability to continuously innovate, sluggish response can break an organization. But defining business requirements for BI doesn't have to be difficult. Let's explore some basic precepts to steer you down the right path.
Defining Business Requirements
Defining business requirements for BI success begins with identifying the key business strategies and objectives at a high level. Taking a top down approach, the next level of managerial information needs in support of strategies and objectives must be determined. This process should continue until the lowest level of information is defined. Remember: requirements define what is needed or desired, not how to do it.
As you consider the business requirements, ensure that they are properly documented, validated, and represent the collective needs of the organization. Also, keep their format simple. To be sustainable over time, the documentation must accommodate the increasingly rapid rate of change and facilitate impact analysis and the adoption of new requirements. Five areas that must be addressed for defining business requirements for BI success are:
1) Performance: These are specific business metrics and KPIs that decision-makers must know to make the business grow and be more successful. Most senior executives rely upon only a handful of metrics to make decisions. Several well-publicized techniques exist to align and define metrics and KPIs in the context of a company's overall business strategy, such as the Balanced Scorecard and the Performance Prism.
2) Dimensional: To provide context to performance metrics, dimensional information must be included. Generally these include time, organization, geography, product or service, account, and customer. For example, currency and time period are dimensions that provide perspective for total revenue as a financial measure. Often referred to as master data, the integrity of these dimensions is critical for consistent reporting across business units.
3) Sourcing: Expectations must be managed when discussing information requirements. Obtaining information that is requested maybe challenging if your organization's systems are not already collecting it. This set of data that must be extracted from source systems in a timely and consistent matter. Sourcing attributes include the definitive sources of the data, specific calculations, data validation, and frequency of extract.
4) Presentation: This refers to the visualization of the performance measures and dimensions required for decision-making. Not all bar graphs are created equal! Experts on interface design argue that depicting qualitative and quantitative information is truly a science, not just an art. What management needs most is the story behind the numbers and in the manner that is most useful to them. What insight should be derived? Do the numbers corroborate other facts? What actions must be taken to improve?
5) Workflow: This represents the sequence of activities decision-makers take based on the values and acceptable thresholds of one or more performance metrics. The way management uses information to make decisions yields a keen understanding of requirements. Thus, usage is an excellent test of need and may form the basis for a business case.
In short, executive-level business requirements provide the linkage between the strategic vision and the key metrics required by decision-makers to manage the desired business outcomes.
Gathering Business Requirements
Once defined, collecting precise facts about what information managers use to make decisions can be gathered. The two most common methods of gathering business requirements for BI are through interviewing decision-makers and analyzing the current systems. The deliverables above should be used to collect, synthesize, and validate decision support needs.
Gathering requirements is about quality, not quantity: asking the right question is 90% of the answer. So, avoid broad questions like: "What do you need?" and focus on more targeted inquiries. If interviewing an executive, a few questions you might ask include:
- What are the most critical priorities for 2005-2006? What metrics are used to evaluate the progress towards addressing or achieving these critical priorities?
- What information do you need today that you currently do not have access to or receive on a timely basis?
- What are the potential barriers that you foresee to achieving these goals? What information is needed to alert you to these barriers so they can be addressed before they become problematic?
Prioritizing Business Requirements
Prioritization provides the options that management must consider when evaluating the information initiatives that will yield the greatest benefit. With finite budgets, this means making tough decisions about the delivery sequence. Logically clustering requirements by benefit and risk is a frequently used option evaluation technique. Benefit refers to ROI, and risk refers to a qualitative or quantitative assessment of business risk, implementation timeframes (time to market), and resource availability and skills. Affinity Analysis and the KJ Method (developed by Dr. Jiro Kawakita) are common prioritization methods, but no single best method of correctly prioritizing requirements exists. Regardless, select one technique and use it consistently.
There's much to be said when it comes to business requirements for BI success. This article focuses only on business requirements; other types, such as technical and operational requirements - which support the delivery of the business requirements - should also be understood before you embark on your journey. But when the time comes, keep this in mind: it doesn't matter how your work is organized or what you call the deliverables. Whatever gets produced must be tangible and actionable. BI is about delivering information on the performance of the business. Tailoring BI to what you need to know can and will dramatically improve business performance.
Mohit Sahgal is a senior principal with the High-Tech and Manufacturing practice at Knightsbridge Solutions. Sahgal has extensive expertise in data warehousing, enterprise information architecture and complex data integration. He can be reached at msahgal@knigthsbridge.com.
For more information on related topics, visit the following channels:
