In my column, Today's Business Intelligence Landscape," DM Review, I discussed the trends that are affecting business intelligence (BI) today. To recap, these trends are: the proliferation of BI analytics into vendor software; the increasing number of standalone BI applications; the increasing independence of businesspeople in the decision process regarding BI; the ongoing practice of many BI vendors to add functionality to broaden their market niches; and the growth of near real-time BI.

The proliferation of analytics that comes from applications with embedded analytics, standalone BI applications and vendor consolidation and/or product expansion challenges the ability of BI organizations to develop a true competency in business analytics. It raises questions about the nature of analytic competence and its importance for a company's success with business intelligence.

The Power of Focused Analytics

Including analytics, particularly online analytical processing (OLAP) capabilities, in software applications provides the power of analytics focused on the application. This is particularly effective with large applications such as enterprise resource planning (ERP), supply chain management (SCM) and customer relationship management (CRM). Analytics provide a basis for understanding data trends and relationships within the application and provide flexible analysis of operational functions such as vendor management and marketing campaign analysis, which these applications support.

By combining analytics with their applications, vendors can offer several advantages. The data stores for these applications are proprietary and complex, and the vendor can provide value to customers with predefined extract, transform and load (ETL) processes, dimensional structures for analysis and queries. These simplify the process for utilizing and analyzing the data used in the application. This makes it easier to get application analytics running more quickly and at less cost than developing a separate BI solution. In addition to lower costs and increased productivity, embedded analytics make things easier for end users. This is another advantage.

However, the integration of applications with analytics has disadvantages as well. The cost and productivity advantage stems from the predefined capabilities provided by the vendor. These are based on the vendor's expectation that the application will be used as delivered or without any modifications, and this is rarely the case. Modifications and customizations can drastically alter cost and productivity advantages of embedding analytics in the application depending on how much the predefined ETL, dimensions and queries must be modified to meet the particular needs of the business.

Also, as analytic modules become a fundamental part of the overall application, vendors are using analytic processes to trigger transaction events. Closed-loop analytics, or turning information into action, is a leading practice for BI and is a natural fit with integrated analytics. However, vendors in general are making the analytics a part of the application process, either as a bridge between application modules or as an initiator of transactions, both of which limit the flexibility of altering the defined analytic structures and activities.

This does not diminish the impact and power of analytics focused on a critical application or business functional area. In fact, many companies have created data marts to provide this kind of support, albeit with the challenge of understanding the source data and developing analytic dimensions and queries without predefined objects to guide the process. In fact, proliferating analytics is really creating data marts tied to applications.

Levels of Competency with Analytics

What does it mean to develop competence with BI in this environment? Architectural practices such as the hub-and-spoke architecture, enterprise information integration and so forth are made difficult by analytics embedded in applications. The expansion of BI products' capabilities and the consolidation of BI vendors make it difficult to establish a BI infrastructure. The accepted wisdom of leading BI practices is being challenged by this proliferation of analytics.

Competency with analytics consists of three factors: the ability to use the analytic tool; effective use of data in the analytic environment; and the ability to adapt the tool and the data to help answer new, unanticipated questions.

Embedded analytics provide a framework to develop competence with the analytic tool integrated into the application (although mastering the tool can be a challenge) and the data used by the application. The first level of analytics competence has now become developing capability with the applications' embedded analytic tools and data. This capability serves business operations and business analysis very well. However, this is not the end of developing analytic competency.

The ability to find answers that help with unanticipated questions is where competency with analytics is truly demonstrated. New questions present a problem when the available data is insufficient for illuminating the situation. This occurs when, for example, CRM data is not consistent with ERP or financial data. The ability to deal with inconsistent and incomplete data is the second level of competence with analytics. This level requires integrating information from disparate sources into a single data warehouse for analysis.

It is integrating data from sources outside the application, with minimal impact on predefined ETL, dimensions and queries, that is a problem for analytics embedded in applications. The separate enterprise data warehouse can source data from analytic environments tied to ERP, SCM and CRM applications or directly from the applications themselves, depending on the level of data granularity required.

The third and highest level of competence with analytics is advanced pattern recognition and data mining. This level is attained by companies and organizations with advanced experience using data analytically. Experience has shown that the ability to perform effective analytics at this level consistently depends upon competence in the first two levels. After all, it always comes down to knowing the data first and then mastering the tools.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access