The 3 Ages of Business Intelligence

Spring 2005

We stand on the cusp of a new era in the evolution of Business Intelligence. Thanks to web and mobile technologies we are rapidly entering the era of the intelligent business network. What will it look like and how did we get here? Let's start back at the beginning to find out...

BI 1.0: Where is our intelligence?

The first Age of Business Intelligence was characterized by production reporting on mainframes.Individual reports were written to specification by expert programmers on behalf of business stakeholders (i.e. Analysts and Senior Management). Depending on the priority and complexity of the report processing of these requests could take days, weeks and even months to execute.

The most enduring concept to emerge from this period was the Executive Information System (EIS). The fundamental technical challenge was the channeling the data from the source systems into a strategic summary. Today this challenge is embodied in the Business or Executive Scorecard. The key difference being the methods of delivery and the business methodologies driving the information displayed on the scorecard.

Focus:

Providing decision support information to the executive layer of the enterprise.

Key Technology Concept:

The Executive Information System

Market Leaders:

SAS and IBM

Change Drivers:

End Users wanted direct access to the data to construct complex queries in 'real-time' as their investigation directed the need.Windows emerged as the 'revolutionary' GUI that facilitated the development of 'WYSIWYG' reporting tools.

BI 2.0: Intelligence in a Box

The second Age is the type of enterprise scale Business Intelligence we see today. It is characterized by end-user friendlier client/server-based BI tools and centralised Data Warehouses (DW) configured to deliver preformatted information to specialist analysts and 'Expert Users' within management. (Thereby removing the requirement for programmers to code individual request)

Market Leaders:

BI Tools: Business Objects, Cognos, and Hyperion. (Emerging: Microsoft) DW: Teradata, Oracle

Focus:

Most of the effort has been on addressing the data issue 'data integration, data quality, data cleansing, data warehouse, data mart, data modelling, data governance, data stewardship'. BI tools are dependent on these efforts.

Key Technology Concepts:

The data cube and the data warehouse

Change Drivers:

The complexity of the current BI Technology has restricted the role of BI to a niche solution for expert users. Spreadsheets remain the reporting and planning tool of choice. For the BI Industry to grow it must evolve to service the estimated 90% of the organization who do not use the existing BI products. - Hence the Microsoft PerformancePoint 2007 Strategy

The division of analytical from operational systems has proven inefficient. It has made it difficult for managers and operational employees to connect analysis with decisions and action and has left open time-consuming and expensive gaps in overall workflow. Plus, current BI practices are weak when dealing with the unexpected and the urgent because of their "data first" orientation.

The 1990's witnessed the emergence of a new 'Blue Collar' worker class (i.e. the Knowledge Worker) whose work day revolves around the endless manipulation of corporate data (i.e. intelligence) in Emails, Word Processes, Spreadsheets and Presentation software (i.e. MS Office). Business leaders have identified the need to automate this activity (much of it 'make work' command and control exercises within organizational silos) to reduce the relatively high fixed 'overhead' costs of doing business today (e.g. HR and Financials).

The WEB has emerged as the 'Revolutionary' collaboration medium.

BI 3.0: The Intelligent Network

There is a growing acceptance of the idea that analysis is a collaborative (not a singular) effort. This means BI must evolve beyond providing a bolt on 'web interface' into an integrated collaborative enterprise network. In this new collaborative world fundamental BI 2.0 concepts, such as "single version of the truth" in data warehousing, are obsolete.

There is also a growing awareness that just being informed by analytics is not enough. If BI is to be of operational value you need to be able to take 'real time' action as and when it is required. To achieve this you need the intelligence embedded in the system.

To meet the challenge BI is evolving to become proactive, real-time, operational, integrated with business processes, and is now extending its reach beyond the boundaries of the organization to other organizational stakeholders such as suppliers, partners, customers, and government agencies to improve information delivery and decision support functionality for all.

In the future BI will find its way to the vast majority of users by being embedded within the operational applications already used by these end users.

These Intelligent Business Networks will initially automate repeatable, operational decisions to address both performance management and compliance issues before evolving into self-regulating intelligent networks proactively monitoring and improving the performance of the enterprise.

This means the current focus on monolithic and expensive BI suites and data warehousing "stacks" will lose favour as organizations go in search of solutions that thrive within the wider, integrated business process network. Business Intelligence 3.0 will become 'operational BI' supported by a technical architecture dominated by intelligent software agents.

BI 3.0 is where the Mobile Convergence (MobCon) revolution infects the enterprise.

Three Radically Different Concepts of Time

Each phase of the evolution of BI represents a very different concept and application of time

BI 1.0 was dominated by the concept of the timeline.

It was taking at least overnight to generate an urgent report. Sometimes it would take days and weeks. If you were last on the priority list it may have taken months to get the report you needed.So the central idea of the timeline required to generate the result of an enquiry was central in the mind of the analyst or executive seeking answers from the mainframe systems.

BI 2.0 is dominated by the cyclical nature of the ETL process.

ETL is the process that gathers data from the outer systems and centralizes it in the data warehouse overnight (or what ever frequency the cycle is set at). So although the analyst has access to the data on their own terms they recognize their investigations are restricted to 'looking backwards into the past' and the relative frequency of the ETL cycle. This is the concept of cyclical time and routine practice.

BI 3.0 will be all about synchronicity

BI 3.0 recognises that if BI is to be of operational value then the organisation need to be able to take 'real time' action as and when it is required. To achieve this you need the intelligence embedded in the system. This is the concept of synchronicity or opportunity.

'It's the ability to go where the puck is heading, not simply follow where it has been.' Wayne Gretzky (Canadian Ice Hockey Legend)