FREE DM Review Site Registration!
Sign-up today and access DM Review on the Web!

Your FREE registration entitles you to:
FREE email newsletters

FREE access to all DM Review content

FREE access to web seminars, resource portals, our white paper library and more!

   

Business Intelligence: The Chicken or The Egg

Business Intelligence has turned 17 years of age as was pointed out in a Computerworld article with Howard Dresner last year. Yet in the article, Mr. Dresner stated that just 36 percent of BI implementations have attained objectives with measurable impact. Howard Dresner's original concept of BI, "ways to deliver information to end users without needing them to be experts in operational research", seems straightforward. How is it that so many of us are trying to explain BI 17 years later? I believe the answer lies in two words used in Mr. Dresner's definition: Deliver and Information.

A premise of BI is a mutual relationship and partnership of IT (deliver) to Business (information). IT is pressured with relentless yearly consolidation and depletion of resources. It is tasked with the responsibility to implement technologies and push the envelope in functions and features, to dispense a growing amount of information faster to an ever-increasing audience. The business user is challenged to employ new business practices, defining rules and standards in support of tactical and strategic objectives. These objectives, often theoretical in nature, are difficult to interpret into specific measurable terms without first establishing standardized disciplines embraced by the leadership chain of command. To further complicate the matter, BI by nature is an evolving concept driven by the competitive market and new technologies to support it. As the business strategy changes so must the BI strategy in order to support and keep pace.

THE CHICKEN OR THE EGG PARABLE

In the chicken or the egg parable we ask which came first, the chicken or the egg? In BI we debate whether technology brings about BI, or whether BI drives technology. To answer this question, let's look at the approach of a business utilizing its resources to compete in the marketplace.


Hailing from the Midwest, I likened the chicken or the egg question to a fictitious enterprise business strategy: produce chickens for resale. The initial strategy or objective on the farm is to successfully hatch chickens and raise them until ready for delivery. The BI strategy is to monitor chick production per hen in order to manage output capacity. To produce chickens, processes are set up to generate fertilized eggs, incubate, hatch, and care for the chicks until they can survive on their own. Components to produce fertilized eggs such as a hen, rooster, feed, water, a nest, daylight, and the correct temperature are defined from these processes. To monitor production results, these elements are designed into a warehouse of data where event activity is collected. The data warehouse will serve as the primary source of performance information. The hen is the key component, used in the egg production, incubation, and the chick brooding processes. From this scenario, we deduce a metric of the number of hatch chicks per hen, which is applied to the business model to produce chickens on an annual basis. The process is shown in Figure 1 (above).

INCREASE PRODUCTION


A new business strategy is established to increase the production of chickens in a given year. The BI Strategy will need to include another metric to compare the new process to the old. Without this metric, there is no way of quantifying the value of increasing the chick production. The new strategy requires modification of the business processes and new infrastructure: artificial lighting, insulated building, additional hens, and artificial insemination to fertilize the eggs. The data warehouse is modified to include the new elements in order to quantify the value of the new process. In Fig. 2 (above), the expanded BI strategy, additional processes and a newly adapted data warehouse are clearly evident with business components synchronized back to the new strategy.

GOING GLOBAL


The success of our new production output generates working capital to expand our business to global markets. This requires redesign of our chick production process, including new technologies and services, in order to meet global demand. A new business strategy includes increase chick production at a lower cost per chick. To minimize risk of undeveloped or failed to hatch eggs, technology replaces the hen in egg production, incubation and brooding. A new BI strategy is instituted, measuring the cost associated to incubate eggs, brood chicks, and monitor the success rate of purchased fertilized eggs. Fig. 3 (above) shows our business in a whole new structure to support mass production.

ANALYSIS

Through the chicken and the egg metaphor, one can see the direct relationship between the data warehouse and the BI strategy. It is clear that technology and the business strategy are bound together. Due to the change in business strategy, the BI strategy and processes needed to evolve. In BI we ask, "What caused the result?" Was it the business process or was it outside the guidelines, business direction or technology? The change in business strategy toward global distribution caused segments of existing processes to become antiquated, no longer able to meet the demand of new objectives. New technologies were acquired to fill the gaps along with a restructuring of the data warehouse. The new warehouse design supported the new BI applications needed to track progress and attainment towards the new objective. The alignment of operational process components with services like egg delivery and reusability of incubators and brooders provided flexibility with the online business processes, thus keeping the system moving around the clock. In essence, to achieve global prominence, a Service-Oriented Architecture (SOA) was put in place to tie all of the production processes together. The building asset was utilized as a reusable service to incubate then manage the chicks until they could be on their own. Egg production was resolved through an outside service, integrated directly into the production process.


Aligning the Key Performance Indicators (KPIs) to the production and distribution processes enabled BI solutions to measure production and determine the maximum growth opportunity. Alert mechanisms were added to the production system to report shortcomings against established business thresholds. These are all factors of integrated Business Performance Management, (BPM), within the framework of an SOA configuration. When BI was coined 17 years ago, technology did not exist to deliver a BPM model consisting of integrated real-time events, loosely coupled into customized business processes. The chicken and the egg metaphor demonstrated how application of evolving technologies to existing processes [filling the gaps] enables the achievement of success as directed by the business strategy. Figure 4 (above) is an illustration that shows the convergence of traditional BI fundamentals, (Fig. 1), to a fully aligned and integrated services based BI architecture carried out in our final objective.


SOA by no means was the intended goal; global distribution was the goal. SOA became a cost effective technology solution to enable success in reaching the goal. Budget objectives were achieved through services and reusable processes utilizing existing assets within the structure. Shawna McAlearney, editor at Application Development Trends, said it best: "You have to understand what you're trying to accomplish before you begin to build. Ask what you have already in your inventory, look for gaps, and that should drive the rest." At each step of the changing business strategy, new BI strategies and business processes were developed before the gaps in technologies had been identified and resolved. I subscribe to the notion that there are five mutually dependant Business Intelligence components to address in a best practice approach to a successful implementation (Fig. 5 above).

  1. Business Strategy provides the foundation for purpose and direction
  2. People interpret and organize the enterprise structure to implement the strategy
  3. Business Processes are the means to deal with the activity that serves the enterprise
  4. Technology enables these processes to function and provide the desired level of service in support of strategy
  5. Data is one of the sources of information to understand enterprise performance

Too often we focus on data or technology and forget the other components. A chicken is first required to produce an egg. We know there are factors beyond the chicken contributing to the egg production. Successful BI strategies are not based on look and feel of technology. BI is a modern means to a basic principle about managing your business as reflected in our metaphor. Insight is not derived from data and technology alone. As Fig. 5 shows, the business component determines how the technology and data are structured and utilized.

WHERE WE GO FROM HERE

The Information in Dresner's BI definition arrives in many forms, all of which provide value to the business strategy. Information is derived from a broad collection of sources; transactional and legacy data, certified (internal and external) services, enhanced (operational) processes, and crucially, from direct experience, i.e. the people factor. To hit a home run requires commitment, accountability, innovative thinking, and dedication from individuals who understand the value and direction of BI to their institution. In this information age it is unreasonable to provide the masses access to anything and everything. Expectations and responsibility need to be clear so that parameters and business rules will govern business processes to insure performance viability and return on business value.

Over the years we have heard about an organization support structure called the BI Center of Excellence (CoE). A group of individuals from IT and business working as one, directly aligned to the executive sponsor, departs dramatically from the traditional organization arrangement. This is not surprising: BI by definition promotes non-traditional, outside the box thinking and approaches to solving business needs. The CoE is the conduit between the individual business units and executive leadership.

The time has long passed where one can afford to ignore technology advances that support an integrated enterprise solution. BI has already reached beyond senior management to essentially everyone in the workforce. Given that only 36 percent of initiatives are a quantifiable success, I believe that technology is far ahead of the ability of the consumer to realize its potential, to implement a state-of-the-art successful plan, or even keep pace with current events.

Before beginning any BI initiative, a readiness assessment of your institution ought to be conducted. Wayne Eckerson, director of research and services at TDWI, includes a Readiness Assessment test in his book, Performance Dashboards - Measuring, Monitoring, and Managing your Business. DecisionPath consulting also has a test available online. Evaluations such as these will ascertain the depth and understanding of the institution's conviction to BI.

A LOOK TO THE FUTURE

All of the industry analysts see Governance as the big push in organizations this year. They also indicate there is a strong desire to expand the BI visualization component, which broadens use on an enterprise level. We have DW 2.0 taking us directly into event driven and in-process analytics from physical and virtual sources of information. Eric Kavanagh, Web Editor at TDWI, recently suggests that new technology will reshape licensing strategies and provide expanded functionality delivered as services. Gartner, Forrester and others are talking about the consolidation of BI Vendors with the BPS vendors. It remains to be seen what impact software as a service, open source, search and the integration of performance management will bring to the new era of BI solutions.

The basic principles of successful best practices still apply. An outstanding document, "Back to Basics," by Greg McMillan in DM Review outlines four pillars of success to achieve excellence in BI:

  • Knowledgeable People have the greatest impact on total cost and degree of success
  • Proven Processes to guide work into the most efficient and high-quality paths
  • Technology that Works: Performance, integration, and usability issues will drive up costs if not functioning as expected
  • Clean Data: Quality of the information sources is critical to the credibility of the process

CONCLUSION

It is clear that technology is not the root cause for the failure rate reported in the Howard Dresner article. Most failings rest on the shoulders of the business consumer who continually focuses on technology and data sources instead of the business strategy, processes, and definitions. Politics and unwillingness to accept the new paradigm, the continued application of traditional approaches and the lack of a cohesive workforce keeps many organizations in the batter's box and others at first base. BI is not about fancy Web-based reports providing statistics from compiled information in a data warehouse. BI is no longer just looking backwards. Technology now enables us to look at what is occurring at this instant and to forecast what may materialize. BI today is about guided decision-making through the delivery of information from a variety of available resources. The concept Howard Dresner raised 17 years ago is now accessible at new levels.

REFERENCE ARTICLES

  • Computerworld: Business Intelligence at Age 17 10/2006
  • Gartner: Magic Quadrant for BI Platforms, 01/2007
  • Computerworld Executive Briefings: Mainstream BI
  • Knightsbridge White Paper: Top 10 Trends in Business Intelligence for 2007
  • CXO America: Will it be Extinction for the Enterprise Software Model? TDWI, 12/2006
  • DM Review: Back to Basics 11/2002
  • DM Review: On-Demand Business Intelligence and Software as a Service: Revolutionizing Small to Medium-Sized Business 01/2007
  • TDWI: Putting the Business Back into BI 04/2006

BOOKS

  • Performance Dashboards - Measuring, Monitoring, and Managing Your Business (WayneW. Eckerson, Wiley, 2005)
  • BI Strategy - How to Create and Document (Michael L. Gonzales and David L. Wells, HandsOn-Press, 2005)

John J. Zeller leads BI coordination at the Mayo Clinic Foundation. He can be contacted at zeller.john@mayo.edu.

For more information on related topics, visit the following channels:



Industry Vendors