Editor's note: This article was originally published at http://www.businessintelligence.com/ex/asp/code.55/xe/article.htm.

Having laid the foundation stone of common metadata in my article on “Common Metadata – The Foundation Stone for Intelligent Business,” I would like to explore the next steps involved in moving toward this kind of setup. As a reminder, in that article I defined intelligent business as taking business intelligence (BI) and putting it at the very heart of the enterprise so that operational applications (executing as part of a business process) and portals can request trusted business intelligence on demand. All this was in addition to the traditional use of business intelligence. However, it is actually much more than this because making intelligent business happen requires the coming together of two main initiatives within the enterprise that are both underpinned by common shared metadata (shared business vocabulary). These two initiatives are integrated business intelligence and enterprise business integration (see Figure 1). 

Figure 1

I have already written at length about integrated business intelligence and common metadata, both of which are fundamental to success. However, we need to also allocate time to this other area – enterprise business integration. Although this may appear to be “off the beaten track” to discuss in the context of business intelligence, it is absolutely fundamental. This is because you have to appreciate the bigger picture of what is going on in an enterprise to really see how BI can be integrated into operational systems by exploiting the enterprise business integration initiative. Equally, people responsible for implementing business integration need to include business intelligence in their strategy.  

At present, in this tough economy, it is enterprise business integration that is dominating IT investment, and there are several solid good business and technical reasons for this. You just have to look at the following example set of symptoms of a fractured enterprise to appreciate why. Lack of integration can cause the following to occur: 

• Confusion over what data means (i.e., no common understanding).
• Annual changing of business objectives even though the goals set for the previous year were never achieved.
• Inability to scale the business due to lack business process integration and automation. This means that fixed costs grow as you try to scale and that too few costs are variable. Hence, operating costs are too high due to high headcount because too many activities are manual. It also means process errors because systems are not integrated.
• Users of one system are not notified of events that occur in other systems.
• No timely response to business events in business process operations (e.g., change order, cancel order, payment overdue, out of inventory). Hence, opportunities are lost and you can’t respond to problems when they arise.
• Rekeying of data into different systems.
• Reconciliatory spreadsheets everywhere flying around the enterprise via email.
• Massive data redundancy and fractured inconsistent data in different systems.
• Conflicting customer treatment by customer service, marketing and sales in different channels.

If you recognise any of these symptoms in your enterprise, you are not alone. I have seen a single changed order event in a business cost $1.8m because of lack of integration and not being able to respond or even know it happened in time to do something about it. Many business executives are plagued by this and are investing in enterprise business integration to deal with it. Integration is therefore dominating because it significantly reduces costs in business operations, improves efficiency and increases self-service. For me, 2003 was the first year that senior businesspeople really recognised the business impact of not having integrated business processes across the enterprise and between enterprises. They realised that they had too many standalone business silos with no integration across systems, applications and processes. In addition, common consistent enterprise business rules often don’t exist, and there is little or no understanding of business events or access to timely information about business problems and opportunities that arise when these business events occur. Pile this on top of fractured inconsistent data with no common data names and business intelligence not available to people in many operational roles, and then you understand the frustration in business. Also, many businesspeople believe that IT complexity supporting these silos is increasing faster than the business value from it and that this complexity is resulting in IT total cost of ownership being much too high. Frankly, I believe they are absolutely right. What is the point of technology if it cannot help a business meet its objectives?

Hence, many companies are integrating to join up their business while simplifying their IT setup and getting rid of IT complexity to get more out of their existing systems (see Figure 2). This IT simplification is like a “corporate Atkins diet.”

Figure 2

Companies are seeking to standardise on IT infrastructure by going with fewer vendors who offer integrated technology platforms. We are seeing this in BI with standard BI platforms (a single end-to-end suite of integrated tools for building and deploying integrated BI) now available from single vendors. On the operational side, we are also seeing single complete business integration platform (technology stack) solutions now available from infrastructure vendors that support a common approach to all internal and external integration of user interfaces, business processes, applications and data. Examples of such integration platforms are shown in Figure 3 and consist of a suite or “stack” of integrated technologies including: 

• Enterprise portal technology with integrated collaboration tools for deploying single secure and personalised user interfaces to content and allowing people to collaborate with each other via tools such as net meetings, threaded discussions, Web chat, email, instant messaging, etc.
• Business process management technologies (process modelling, simulation, execution and monitoring) to integrate and manage integrated business processes across and beyond the enterprise
• A common shared rules service
• An application server
• Application integration technology
• Data integration technology for on-demand data integration targeted at applications

Figure 3

You might ask what this all has to do with intelligent business. The answer is simple. If a common integration platform is being used to integrate operational business processes across the enterprise, then trusted business intelligence can be plugged into this platform to maximise the opportunity of creating an intelligent business. This is especially true if a common business vocabulary (common metadata) can be shared across both a business integration platform and a business intelligence platform to put everything on the same footing. In this way, the user sees all operational data and business intelligence in a shared business vocabulary via their single secure personalised portal user interface.

Hence, the intelligent business must include key technology components from both the operational and analytical worlds. These components are:

• A standard business intelligence platform
• A standard enterprise business integration platform
• A shared business vocabulary via metadata management and metadata integration
• Corporate performance management software integrated with analytic applications sitting on top of the business intelligence platform and integrated with business processes

Stages to Implementing Intelligent Business
It follows that building the intelligent business requires that work be done in both operational and BI areas to prepare the ground to then combine business intelligence with operational business processes. Therefore, there are four main parts to implementing an intelligent business strategy. These are:

1. Define an intelligent business architecture
2. Integrated business intelligence (the BI element)
3. Enterprise business integration (the operational element)
4. Intelligent business (the combining of operational and BI)

Note that parts 2 and 3 can, and should, be done in parallel. Let’s look at these stages in more detail.

The first stage is to create an intelligent business architecture that acts as a blueprint for implementing the intelligent business. This architecture is effectively an enterprise architecture that brings the worlds of operational and analytic applications together in the context of business processes and that presents content to users via a personalised, secure Web-based user interface – an enterprise portal.

Integrated Business Intelligence (the BI element)
Integrated business intelligence involves a number of things. Perhaps the first is considering whether or not to standardise on a common BI development platform from a single vendor to reduce the complexity of the BI environment and to act as a common suite of technology for BI development. Vendors such as Business Objects, Cognos, IBM, Microsoft, Oracle and SAS are all offering business intelligence platforms. While this might seem a step back for companies with mature BI setups, it is a strategic move to simplify technology and focus skills on a single technology suite. Currently, I have a number of clients looking to do this, which I think is fairly representative of the industry at large.

This stage also involves repairing existing business intelligence systems by implementing/enforcing a shared business vocabulary (data names and definitions) across all custom and packaged business intelligence applications. The reason for doing this is obvious – common understanding and consistency. Having done this, business intelligence can be integrated such that any islands of custom and package-based BI applications and tools within the enterprise can come together and also be integrated with corporate performance management (CPM) scorecards and dashboards for rock solid enterprise corporate governance. This involves creating metrics trees in CPM software that link scorecard KPIs to metrics in existing underlying BI applications. Also, business objectives and objective “networks” defined in corporate performance management software can be shared across the enterprise by integrating corporate performance management with enterprise portals so as to create an enterprise-wide performance culture. In this way, people understand what their objectives are and how they tie back to common strategic objectives so that they know how they are contributing to strategic goals. In addition, BI platforms can be integrated with collaboration tools so that business users can collaborate and share intelligence with others before making joint decisions.

For operational performance management, companies need to capture operational events by integrating data integration software (sometimes called extract, transform and load/ETL software) from the BI platform with application integration platforms (sometimes called enterprise application integration/EAI software) so as to collect events and trigger on-demand near real-time data integration of structured and unstructured content when specific events occur. In addition, adding event-driven on-demand automated analysis and using a common rules engine to implement rule-driven alerting, recommending and action messages will provide support for business activity monitoring, on-demand recommendations and alerts for guiding operations personnel toward achieving business objectives in the intelligent business.

Enterprise Business Integration (the operational element)
Enterprise business integration should ideally be done alongside integrated BI so that teams in both areas work together and plan where and how to integrate business intelligence into operations. Based on this architecture and achieving specific strategic objectives, this is about implementing integration at five main levels:

1. User interface integration
2. People integration
3. Business process integration
4. Application integration
5. Data integration

so that these integration initiatives are done in unison to achieve a common strategic business objective (e.g., to reduce operational cost).

User interface integration is done using enterprise portals. Enterprise portal technology allows a company to deploy role-based portals to give customers, suppliers, partners and employees personalised access to integrated content.

People integration is achieved using collaboration tools such as email, presence awareness, instant messaging, net meetings, shared calendar, etc.

Business process integration involves the separation of business processes from applications so that integrated business processes can be separately modelled (using business process modelling tools) to guide the execution of business process acrossmultiple applications inside and outside the enterprise. Business process modelling tools allow companies to design and redesign business processes and then generate an XML “instruction set” in a standard XML language called business process execution language for Web services (BPEL4WS). Examples of business process modelling tools include BEA AquaLogic Business Service Interaction, IBM WebSphere Business Modeller, Biztalk Orchestrator, IDS Scheer ARIS and webMethods Modeller. The BPEL4WS document generated by the modelling tool goes into a process engine that then parses the BPEL and issues messages to the appropriate applications and people to carry out the business activities in the process (e.g., a credit check) in the order specified. This is shown in Figure 4. Once underway, executing business processes can then be monitored using business activity monitoring (BAM) and activity-based costing (ABC).

Recent articles by Mike Ferguson

• Web 2.0 and Business Intelligence – How Do They Fit Together? Part 4
• A Business Process and Performance Management Framework for the Intelligent Business