BeyeNETWORK - The Global Vision for BI and Beyond

Over the past several years, there has been widespread adoption of business intelligence (BI) and data integration strategies. Industries that are focused on efficiency and making fact-based decisions have increased the demand for today’s enterprise analytics software that allows users to view, manipulate and question the data in front of them with unprecedented ease and speed. Decision makers can now have the most up-to-date insight into their company’s data, allowing them to not only ask questions of their data, but to have answers instantaneously.

When companies can utilize these powerful tools to enhance their decision-making abilities, why then are the vast majority of decisions still being made without the benefit of these tools? Decisions made in the company’s “trenches” are often based on experience and intuition, rather than hard facts. With millions of knowledge workers spread throughout an organization, the ones actually driving the business’s performance often place their bets based on their guts because today’s business intelligence only serves the small population of decision makers residing in the ivory towers, those placing the company’s “big bets.”

Not anymore. The next generation of BI is here, and to say that analytics packages are becoming more “user friendly” is an understatement. Today’s analytics packages, through enhanced user friendly capabilities, can now be used to increase decision-making power in all facets of the company from sales to marketing to R & D.

With such increased accessibility and insight into the information by all the decision makers, a new trend is emerging – the fact-driven enterprise. By giving more users access to data through the use of analytical tools, companies are able to increase performance. With the ease and speed at which workers can now access, understand, question and re-analyze information, companies can improve their decision-making processes based on facts, rather than hunches and gut reactions.

These next-generation tools are providing employees at all levels within organizations the ability to make fact-based decisions. Today, companies are moving the information provided by analytics out of the board room and into the front office, where the vast majority of a business’s decisions are often made. By bringing these analytical capabilities, previously only available to the C-suite executives, to all levels inside an organization, everyone now has insight into information as well as the ability to interact with it in real-time and answer questions on the spot.

Examining the Analytics Players

Because of the importance being placed on analytics, it’s no wonder that the noise from technology vendors can be deafening. The sheer volume of players can complicate the technology selection process and mask the best practices around building and deploying analytics in the enterprise. “Pervasive analytics,” “prescriptive analytics,” “predictive analytics,” and “business analytics” all have the same stated benefits, so what makes them different? To make the right selection, it’s important to understand the various technology options available and be able to separate the wheat from the chaff when it comes to how vendors market themselves.

Instead of focusing on marketing terminology, more light may be shed by breaking down the categories of vendors vying for the analytics pole position. These categories include statistics vendors, vertical application vendors, business intelligence (BI) vendors and visual analytics vendors. In the end, most organizations will likely find that some combination of these approaches will be optimal, as no one approach can solve all needs. Each approach has its strengths and limitations, as does each vendor.

Statistics vendors. The statistics vendors have been in the analysis business for the longest amount of time. Products such as those from SAS, SPSS and the open source program are well-suited for analysts and statisticians who are proficient in mathematical methods that generate models for classification, segmentation, forecasting and propensity scoring, as well as the detection of patterns and anomalies. With data available and a specific question to answer or hypothesis to test, statistics packages are extremely powerful. However, the major limitation is the relatively small population of users that can use them because they typically require users to have a statistics or quantitative background. Most organizations use it to serve as a back-office function, answering questions that arise from business users with domain knowledge. While sufficient for supporting planned or big decisions, the typical question-answer cycle and turnaround delays limit the widespread deployment of statistics at this point in time.

Vendors in the statistics category are aware of their usability limitations, as many have attempted to make their interfaces more user-friendly. Most notably, SAS’ JMP product simplifies statistics for novice users, however includes minimal analytics needed by experts.

Application vendors. Broader application vendors, not willing to cede any ground in the analytics race, and not motivated to have users extract data for external analysis, have quickly started including analytics in their offerings. Taking advantage of a large installed base and mountains of data ready to be analyzed, application vendors now offer analytics capabilities on top of their core solutions such as ERP and CRM. For instance, Siebel Business Analytics, now part of Oracle’s BI offerings, provides businesses with pre-built analytics applications in a number of specific functional areas and industries.

For users of the application vendors’ solutions, it is certainly beneficial to have some analytics capabilities available for rapid deployment. As long as the topics for analysis are supported by business functions and data stored in the application, this approach may be beneficial. Users may already be familiar with the interfaces and pre-configured database schemas, ETL scripts, dashboards and existing business practices may be quickly leveraged for deployment under this scenario.

However, this approach is not without its limitations. Application vendors provide business users with limited analytics power. Their approach more closely resembles reports or executive dashboards, rather than true interactive and free-form business analysis. This is partially driven by the system architectures upon which they are built.

Business intelligence (BI) vendors. BI companies form the third category of vendors supporting the analytics space. In recent years, BI vendors have attacked most major companies in the world and have flooded them with countless reports, portals and data warehouses. Their market penetration has been impressive; however, a wave of consolidation led by the big platform players has swept the industry, ending this period of unchallenged growth. As a result, BI vendors are attempting to quickly add analytics as a way to broaden their market caps, differentiate their offerings from other BI competitors and push beyond serving reports in an effort to get closer to front line business users.

Today, most of the pure-play BI vendors, as well as the platform vendors, provide comprehensive BI solutions ranging from ETL, reporting, ad hoc query, OLAP (online analytic processing) and dashboards, while some offer pre-built analytics applications focused on specific vertical and horizontal areas.

BI products have done a good job providing reporting capabilities to many different business users, perceiving them as consumers of “reports.” In addition, they provide OLAP-based analysis for a fraction of business users who want to do analysis. Taking advantage of their vendor-neutral position and comprehensive solutions in BI, some pure-play vendors such as Business Objects provide pre-built analytics applications. Others including SAP, Oracle and Siebel also include ETL, ERP and CRM data to populate data warehouses, allowing business users to access to the data via pre-built reports and dashboards. Businesses that want to provide business users with snapshots of their business and assumable analysis paths with OLAP cubes will find those solutions beneficial.

Today’s BI products do not provide a complete solutions if your organization wants to become an analytics competitor. In fact, you need to provide front-line business users the tools that will allow them to unleash their analytics power rather than reports with static summaries of historical data of traditional BI software that can’t keep pace in a dynamic business environment.

A growing trend in the industry is the commoditization of BI functions into core platforms, such as Microsoft, Oracle and IBM. As these big vendors include more and more traditional BI functions into the database, differentiation for the pure-play BI vendors is eroding. Microsoft, for example, acquired ProClarity for a small investment that will pay big dividends for customers – giving them a front-end BI product to leverage the SQL Server Analysis Services and Reporting Services engines. The acquisition also positions Microsoft to compete head-to-head against the leading BI pure-play vendors. Additionally, it would be beneficial for businesses to employ BI offered by Microsoft because of ubiquity of Excel and less expensive expenditure for the system. As a result, it is likely that there will be more consolidations of vendors and decreased prices of offerings, especially with the emergence of open source BI.

Predictive analysis vendors. Predictive analytics, more commonly referred to as data mining, is yet another category of vendors creating a model to automate the filter and application of the data. In its purest form, predictive analytics is a discovery process supporting the analytics space.

This three-tiered process includes identifying data, finding hidden patterns in past or existing data, and applying that information in order to evaluate future behavior. Because we live in an information-rich age, this effort can play an integral role in a business’ success or failure. There are many roads from commencement to execution, but the path you choose will have everything to do with the final result.

Visual data analysis vendors. The last category of analytics vendors is visual data analytics. Largely influenced by data visualization approaches common in research and development and academic applications, this group of vendors intends to deliver easy-to-use interfaces to bring analytics to a broader base of users across the organization, which just might be what’s required.

While not new, the visual data analysis industry has been historically focused on analyzing data for more technical research applications. Success has been driven by their ability to make sense of volumes of data and to make insights and answers more obvious. Now, several of these vendors are applying their experience to solve problems faced by business users. Attracted by potential profits of cracking the problem of widespread analytics deployment, investors have funded a number of upstarts in recent years.

One such vendor is Spotfire, which provides visual and interactive analytics tools that maximize the value of enterprise BI. Spotfire customers have been early adopters of the newly coined “decision-centric business intelligence” – in some circles also known as BI 2.0. This next generation of business intelligence moves users beyond reporting to include the contextual delivery of business data at the point a business decision must be made within a business process.
 
Spotfire’s applications allow users to visually analyze and interact with data to speed decision making and gain deeper insight. Users are not confined to predetermined reports and queries, but are free to explore information in any way they want. Spotfire’s business analytics applications provide rich graphics and instant visualization into complex relationships and unseen patterns that can provide the keys to competitive advantage. Its products automatically capture and store analytic workflows, letting users publish findings and best practices to libraries, making it easy to share processes and insights with colleagues. Distinguished by its speed to insight and adaptability to specific business challenges, Spotfire’s software rapidly reveals unseen threats and new opportunities, creating significant economic value.

A relative newcomer on the analytics scene is Tableau Software. Tableau provides a desktop-based reporting and analysis tool that allows users to do OLAP-type analysis using cross tables and other graphs. However, it is limited by providing access only to a single data source, which does not fulfill most enterprise requirements.

Other vendors in the space include Visokio and Panopticon, both of which offer a visual approach to data analysis. Visokio, formerly Iokio, markets two main products, OmniScope and FeatureFinder. OmniScope, its primary product, provides business users with capabilities of visualizing data, dynamic filtering, image handling with data records, and publishing data.

The promise of these visual analysis approaches has not gone unnoticed. BI vendors, struggling to reestablish differentiation, have attempted to expand their offerings in the visual analytics space. For example, Business Objects acquired Xcelsius in 2005 and added it to its Crystal product line. But although Business Objects positions Crystal Xcelsius as an analytics tool, it acts just as a more visual report generator. Perhaps not sure of the analytics footrace, Hyperion chose to partner with the newly formed Tableau as their approach to complement their product line and for delivering analytics to users.

Your Road Ahead

There are a number of different approaches your organization can take to effectively be more competitive using analytics. While each approach may range in statistics, vertical applications, BI and visual analytics, each has its inherent strengths and limitations.

Today’s analytics packages now offer user-friendly capabilities, allowing users to increase decision-making power in all facets of the company. With increased accessibility and insight into the information by all the decision makers, any organization can become a fact-driven enterprise.