In the post Three Tiers for Measuring Business Intelligence & Analytic Performance I described my views that Business Intelligence (BI) & Analytics contribute to organizational performance with three different levels of importance.

The primary purpose of BI & Analytic technology is to support decision-making and performance management. The business benefit is derived from the users effectively and efficiently using the technology. There are a few business models where BI & analytics are at the core of the business strategy such as companies that collect and analyze data (e.g. Neilsen), companies that perform analytics for their customers (like ariba), and nowadays, cloud BI vendors or Software as a Service vendors (e.g., Gooddata or Tribologik(r) Oil Analysis). In these cases, BI and analytics are at the center of the business used to derive value. For these types of organizations, BI & analytic performance should be measured using the tier one measures like I described in my last post. The tier two measures will most likely not be a true measure of the BI & analytic value.

After tier one measures, the next best way to measure BI & analytic performance is based upon user or organizational productivity — tier two measures.

In 2011, Dr Thomas Quesssier, Thomas Goettsche, and I performed research to understand how people use of technology in decision making.  The research categorized the software users and the types of decisions that they make using software.  From that study we can see that:

• All size organizations from every geography use some type of BI in decision-making
• BI and analytic solutions are being integrated in operational systems and processes and trigger automated decisions
• All levels of the organization and external parties use BI and analytics
• Board of Directors and Chief Executive Officers are hands-on users (direct use) or receive the results from someone else in the organization (indirect use)
• BI and analytic results are shared with customers and suppliers.

Why are they using BI and analytics

The conclusion from the study is that “BI is used to increase performance, according to respondent reports, by making or assisting real-time decisions, assisting in strategic planning, improving predictions, and contributing to making quantitatively driven decisions.”[1]

For example, the software is used to provide input to the staff who makes the decisions or the technology is integrated into systems or processes that make the decisions. The subjects listed below and shown in Figure 2 are topics where BI and analytic solutions are used.

• Allocation or reallocation of responsibilities or resources
• Forecasting or predicting the most promising strategic options
• Customer mix, competitive emphasis or organizational structure
• Setting the financial land non-financial goals of the organization
• Implementing annual Objectives or short-term strategies
• Individual business unit objectifies and strategies

In the study, we confirmed a positive correlation between BI use and financial performance. Some example cases from the study that highlight this point include:

An United States financial services company that used a financial planning and analysis solution to maintain earnings before tax and interest in the mid 20 percent range and increase revenue by 250%.

An Iranian agency selling cable and related technology was able to add capabilities to predict material cost, transportation and delivery, exchange rate, and other values before making offer. The result was higher profits and improved coordination with suppliers.

A Kenyaian insurance company was able to deploy BI capabilities to its agents resulting in a being able to close policies within the same day and at the point of sale. They were also able to move into new business markets such as micro-insurance.

A “New Europe” telecommunication company that deployed data mining technology for a wide selection of customer and product decisions, including customer profiling and tariff simulations. The result was a reduction in decision cycles from weeks to days.

Thus, we proposed a value chain linking “use of BI technology” with “better decisions” resulting in “better organizational and financial performance”. (Figure 3 Decision Value Chain).

 

Figure 3 Decision Value Chain

With the results of the study in the background it logical to measure the value of BI & analytical performance on the productivity of the users in their day-to-day decision-making and performance management.  The following are some of themes that can be used to measure if BI and analytics are supportIng the users. Examples of tier 2 measures are provided in Table 1.

Decision-making is dynamic. From one day or hour to the next the type of information needed can vary dramatically. In fact, a TDWI report found that reporting requirement change at least monthly for a large percentage of respondents while the data warehouse does not contain the data needed for those reports for nearly a third of the respondents. Other reasons include:

• Large number of data sources
• Growing volumes of data
• Need for structured and unstructured data
• Decision time windows are shrinking
• Lack of skilled resources
• and, and and …

To improve BI & analytics impact on the User Productivity a selection of strategies must be employed.  The following are some ideas that I recommend based upon our research, analyst input, my experience, and different case studies.

• Engage management…we are in fact talking a culture of fact-based decisions.  Incremental changes can occur without management seeing BI and analytics as a core competency, but much more can happen when they are fully engaged
• Establish an entity responsible for supporting decision-making; our research has shown that organizations with a BI compentency center (or similar) have a higher correlation between BI and improved organizational performance
• Build communities of practice that share experiences and ideas on using BI & analytics
• Automate and standardize the creation of routine reports and information
• Provide a selection of different reporting and analytical tools that fit to the different type of user profiles
• Provide options and different environments for working with different types, quality, and volumes of data
• Provide options for different user profiles to have access to different levels of computing power
• Use agile methods for developing BI and analytical solutions
• Consider SaaS solutions. support and encourage users to find niche applications that serve their need. help them with interfaces and agreements– be sure you will own the data.
• Employ some new technologies e.g., visual analytics, in-memory analytics, data warehouse appliances

1. Miller, Gloria J., Queisser, Thomas, Goettsche, Thomas. Mobilizing Technology and Teams for Better Decision Making, May 2011
2. White, David. Agile BI: Three Steps to Analytic Heaven. AbeerdeenGroup, April 2011
3. White, David and York, Matthew. Agile or Fragile? Your Analytics, Your Choice. AbeerdeenGroup, July 2012
4. Stoddler, David. Achieving Greater Agility with Business Intelligence. TDWI Best practices Report, TDWI Research, First Quarter 2013.