The term advanced analytics is gaining ground in vendor marketing. Hardly a briefing goes by without a vendor introducing this phrase into the presentation to try to differentiate itself from other business intelligence (BI) solutions.
At a recent Boulder BI Brain Trust (BBBT) event, a vendor caused a flurry of analyst conversations on Twitter concerning the use of this term in the presentation. Many felt advanced analytics was a confusing term that is abused and overused by vendors. We wondered what “non-advanced analytics” are! Some people even commented that BI had become so broad as to be meaningless.
In this article, we share our views on this terminology. Our aim is not to provide a detailed theoretical or academic treatise, but instead to hopefully start some constructive dialog leading to consensus.
The Value of Terminology
It is important to first address the merit of debating the meaning of technology terms. Most people would agree that it is far more important to focus on the business benefits offered by a technology. Technologies evolve and so do their definitions. At the same time, however, technology terms are used in vendor marketing campaigns, analyst reports and education classes. It is important, therefore, that some level of common understanding exists in the industry about the meaning of a term if we are to avoid confusion and explain how a technology can provide business benefit.
What is Analysis and What are Analytics?
Scanning dictionaries and the Internet for a definition of analysis
didn’t provide us with any useful input to the discussion. Webster’s defines analysis as “The separation of a whole into its components parts,” certainly not a definition that sheds much light.
More meaningful (but still confusing) results are obtained when prefixing analysis with data
. Wikipedia, for examples defines data analysis
as “… a process of inspecting, cleaning, transforming, and modeling data with the goal of highlighting useful information, suggesting conclusions, and supporting decision making.” On the other hand, business analysis
is defined as: “The discipline of identifying business needs and determining solutions to business problems. Solutions often include a systems development component, but may also consist of process improvement or organizational change or strategic planning and policy development.” These two definitions may explain why the IT and business views of analysis sometimes differ. IT often defines data analysis as covering the complete information life cycle from cleaning and transforming source data making it ready for analysis, to analyzing the transformed data and creating analytics. Business users, on the other hand, view business analysis as a set of techniques for defining analyses and creating analytics on the transformed data.
Definitions for analytics, metrics, measurements, and indicators offer a confusing set of options. It is worth noting that analytics
is more often defined as “the science of analysis” rather than the “results of analytical processing.” Given that business intelligence is about providing business users with intelligence about the business, we offer the following:
“Business analysis is the process of analyzing trusted data with the goal of highlighting useful information, supporting decision making, suggesting solutions to business problems, and improving business processes. A business intelligence environment helps organizations and business users move from manual to automated business analysis. Important results from business analysis include historical, current and predictive metrics, and indicators of business performance. These results are often called analytics.”
What About Advanced Analytics?
Analytical processing solutions continue to evolve and change as new technologies and new types of data are introduced into organizations. The result is that companies are at different levels of maturity in their use of analytics. Some may be using basic dashboards of performance indicators, whereas others may have expanded into using more sophisticated techniques such as data mining and predictive analytics. The term, advanced analytics
, is often associated with the more sophisticated capabilities. Vendors who use this term must provide very clear explanations and examples of the types of advanced analytics they support since there are a growing number of capabilities in this area.
Has the Term Business Intelligence Lost Its Meaning?
Although there is no doubt that the term business intelligence continues to have significant meaning in the IT industry, from a marketing perspective the term has become problematic for several reasons.
- It is closely associated with data warehousing. Some business users see data warehousing as an IT technology solution without clear business objectives and goals. This may be why vendors often use the term analytics rather than business intelligence in discussions with business units because analytics are not so strongly associated with technology solutions and the IT department.
- Only a small percentage of an organization’s information (about 20% by some analyst estimates) is stored in a data warehouse. Some 80% of data in an enterprise is considered an untapped resource for supporting decision making. Whereas some of this data certainly belongs in a data warehouse, it is not always practical or cost effective to store all of it in a data warehouse. This is especially true for high-volume event data and most unstructured content (many types of text documents, multimedia content, and content produced by collaborative and office computing, for example).
The result is that certain types of analytical solutions are being built outside of the traditional BI environment. Examples include web analytics, content analytics, and event analytics (CEP-related solutions, for example) built on in-motion data. Some experts argue that this analytical processing and the data used in its processing should be brought into the data warehousing and BI environment. As already pointed out, this is not always feasible. This does not exclude, however, the storing of analytical results or filtered subsets of the source data from being consolidated into a data warehouse.
Given this situation, it is important to realize that although the BI environment provides a mature set of processing capabilities for analyzing historical and low latency structured data from core business processes, it is only one component of a solution for supporting all business decision making. The issue here is that many vendors and organizations think that the BI environment should grow and subsume other technologies associated with decision making. Unfortunately, this approach has severe limitations given today’s BI technology. Instead, the BI environment should be able to interact with other technology components to form a more complete decision support environment. We have used the term decision intelligence
in this context in the hopes of overcoming this mind-set.
In this article, we have looked at a variety of different terms and offered some suggestions for the ways these terms can be defined. Certainly, vendors (as well as consultants and analysts) always have, and always will, create and use new terms to differentiate and help sell their products and services. We suggest that we all become more vigilant in supplying definitions and usage cases to ensure a complete understanding of our meanings. Otherwise, as technologies mature and gain traction in the market, these terms become overused and abused.
For users of decision support technologies, business benefit has to be the main focus, and it is important that technology usage focus on achieving this benefit. Next month, we will discuss how organizations can grow and evolve their decision support environment to increase the scope of business issues being addressed.
Meanwhile, we are interested in getting other views on this topic. Please either enter your comments here or send them to us directly.
Recent articles by Claudia Imhoff, Colin White