Analytic applications are filling a need in enabling organizations to measure, analyze and optimize business performance. IDC created the analytic application category in 1997 to call attention to application software that supported activities beyond the scope of transactional systems such as ERP. Today, analytic applications represent a market that is approaching $3 billion in total worldwide software revenue.
Analytic Applications and Business Intelligence
It is important not to confuse analytic applications with business intelligence technologies or tools. There is a long- standing market for business intelligence tools such as OLAP, query/reporting and data mining. Analytic applications incorporate such technologies, but are fundamentally different in terms of specialization, segmentation and structure.
Analytic applications are specialized to a particular business process or function, while business intelligence tools are generic.
Analytic applications can be segmented by business function (such as finance or marketing), while business intelligence tools can be segmented by technology (such as data mining or OLAP).
Analytic applications structure and coordinate business activities to achieve a particular result (such as producing a budget or assessing the performance of key suppliers), while business intelligence tools can support ad hoc query and analysis that is not predefined.
Analytic applications expand the reach of business intelligence to an extended user base, packaging these technologies in a business context; but analytic applications will not displace business intelligence tools. There will continue to be a need for ad hoc analysis to explore new types of issues and questions as they arise.
Build Versus Buy
Should companies build or buy analytic applications? This is not an "all or nothing" question. Most companies will adopt a strategy of selectively building and buying applications, depending on their business needs, IT skill and application availability.
Companies may choose to adopt different strategies (build versus buy) according to application type:
Financial/Business Performance Management:
Analytic applications of this type measure and optimize financial performance (e.g., budgeting) and/or evaluate an enterprise business strategy (e.g., balanced scorecard). Budgeting and financial consolidation were among the first packaged analytic applications because the processes involved are well understood and because they apply across all industries. There is a move toward integrated suites comprising budgeting and planning, consolidation, activity-based management and balanced scorecard.
Analytic applications of this type measure and optimize the production and delivery of a business's products and services (e.g., demand planning, workforce optimization and healthcare outcomes analysis). These applications differ from industry to industry. Hot applications include demand planning and pricing optimization, and the first packaged applications for optimizing biotech-specific processes are now beginning to appear on the market.
Analytic applications of this type measure and optimize customer relationships (e.g., customer profitability, customer retention and marketing analysis). Customer-specific analytic applications first gained a foothold in the marketing department, but they are now becoming cross-functional as optimizing customer relationships becomes an enterprise imperative.
The Buy Market by Analytic Application
Which types of packaged analytic applications are organizations most likely to buy? Figure 1 shows IDC's estimate of the current market size and forecast for the three major categories of packaged analytic applications. The fastest growing sector is CRM analysis applications, but operations/production and financial/BPM are each significantly larger in the current marketplace. (For more information, see the IDC report "Analytic Applications Market Forecast and Analysis: 2001-2005.")
Figure 1: Buy: Worldwide Software Revenues for Packaged Analytic Applications
Source: IDC, 2002
Though the "buy" market is gaining momentum, IDC believes that companies will continue to build applications for the foreseeable future. One factor is lack of availability; some vertical industries are not well served by analytic applications. A second reason is the strategy to develop unique applications that provide competitive differentiation.
The Build Market by Analytic Application
Which types of custom analytic applications are organizations most likely to build? Figure 2 shows IDC's estimate for the current state and forecasted future state for custom analytic applications. IDC sizes the worldwide market for data warehousing tools. These include software to populate a data warehouse, store the data in a DBMS and access the data using business intelligence tools. Then IDC segments this market for tools by application based on surveys that show how companies prioritize the types of access to a data warehouse.
Figure 2: Build: Allocation of Worldwide Software Revenues for Data Warehousing Tools by Application
Source: IDC, 2002
Financial analytic applications are not nearly as important for custom analytic applications since this is the sector best served by packaged applications. Though CRM is important, IDC believes the operations/production sector comprises the largest part of the custom market. It is in this sector that companies are seeking to drive greater efficiencies in their supply chain, research and development, and other product and service-related functions. (For more information, see the IDC report "Build versus Buy for Analytic Solutions Market Forecast and Analysis, 2000-2005.")
Linking Analytics to Business Operations: The Key to ROI
Whether analytic applications are built or bought, the feedback from analytics must lead to corrective action that impacts business operations. If it does not, there is no clear way to measure its impact on the business.
This implies that analytic applications must do more than provide information. They must guide the decision-making process, leading to actions that improve business performance. To maximize their business impact, analytic applications must evolve as follows:
- Eliminate the disconnect between analyst and decision- maker.
Decision support implies a technical end user bringing together information for the benefit of a manager who has responsibility to make decisions. In this scenario, the actual user of the software is not empowered to make decisions. How likely is it that the information will be relevant? Decision-makers, not just technical analysts, must be direct users of the analytic applications.
- Focus on teams, not just the individual.
Improving decision making requires more than correcting for individual bias. Many decisions are the result of a collaboration between individuals, and biases occur in the dynamics of group interaction. Guiding decision processes requires collaborative support.
- Track the decisions and evaluate their effectiveness.
It is ironic that business intelligence software traditionally has not been interested in capturing decisions. We improve by learning from our mistakes. Analytic applications will evolve to track the outcome of the decisions and measure their effectiveness.
- Capture the decision-making processes of the best performers.
The best performers in an organization use information to make better quality decisions, such as a sales representative qualifying leads or a customer service specialist handling difficult cases. Capturing decision-making processes addresses the fundamental goal of knowledge management: preserving employee expertise. Process support is one of the defining characteristics of analytic applications.
Improving the decision-making process requires supplying relevant feedback that can correct for the irrational biases of the decision-maker. This is the future direction of analytic applications – one that is likely to bring greater business impact and more rapid return on investment.
Henry Morris is the group vice president and general manager for for IDC's Integration, Development and Applications Strategies (IDeAS) solution research group. Dr. Morris started the Analytics and Data Warehousing research service at ICD and coined the term "analytic applications" in 1997. Currently, he is exploring the relationship between business intelligence and business process automation. Morris may be contacted at firstname.lastname@example.org.