Embedded BI is a largely invisible part of the BI market, but for some vendors, such as Logi Analytics, Jaspersoft, and Pentaho, it represents a large portion of their annual revenues. And that revenue stream is destined to grow as new cloud and big data applications embed interactive reporting and analysis capabilities.
As I said in my previous blog ("Historical BI Cycles: From Reporting to Analysis and the Future"), the next major wave in BI is embedded BI. That wave won't become mainstream until we move past the current analysis phase sometime in 2015 at the earliest. But in many ways, embedded BI represents the fulfillment of BI's promise, in which it becomes an information service that informs day-to-day operations and automatically triggers decisions and actions that drive the business.
Karl Van den Bergh, vice president of products and alliances at Jaspersoft, has crafted a model of the evolution of embedded BI. (See figure 1.) Since I found it instructive, I thought you might too.
Figure 1. Evolution of Embedded BI (click to enlarge)
The first generation of embedded BI is represented by Crystal Reports, which was widely embedded in many software products, including Microsoft Visual Basic, usually with a restricted license for up to five users or a single data source. Crystal Reports was proprietary software that developers embedded into their own solutions using code-specific libraries. In the 1990s, embedded report writers, such as Crystal Reports, generally produced static reports.
The second generation in the 2000s marked the ascendance of open source or low-cost, developer-friendly BI products that support interactive reporting, dashboarding, and OLAP functionality. The leading embeddable BI products are Logi Analytics, Pentaho, Jaspersoft, and Actuate BIRT. Developers use desktop tools to create Web-based BI content which they can then embed into host applications using Web services (e.g. SOAP) or HTML tags, such as iFrame. The Web-enabled BI applications run as separate Web sites and thus share no code or libraries with the host applications they support so there are no mutual dependencies or versioning considerations. They can even run on a separate server to avoid performance issues.
Embedded BI has evolved in lockstep with application developers, moving from the desktop to the Web to the cloud. Along the way, BI tools have gotten more powerful and easier to use and embed. They offer application developers the full range of BI functionality--from reports and dashboards to analysis, discovery, and authoring--within an easy-to-use, point-and-click development interface. Moreover, thanks the openness of the Web, BI applications are no longer embedded within the code of a host application, but rather sit adjacent to that application, providing easier deployment and administration and higher degrees of scalability and performance.
Posted May 23, 2013 9:29 AM
Permalink | No Comments |