Embedded Analytics Going (and Growing) Strong
A new survey sheds light on how and why enterprises are using embedded analytics, and shows what developers say is ahead for the technology.
- By James E. Powell
- April 11, 2016
If there's one thing the fourth annual "State of Embedded Analytics Report" proves, it's that embedded business intelligence and analytics is hot. Enterprises consider it a strategic initiative that is often driven by company CEOs or executives, not IT, the report from Logi Analytics reveals, which also notes that "embedding analytics into users' preferred applications provides an instant way to get information into the hands of more people within the context of everyday work."
It's definitely hot for application providers because it adds value. In fact, two-thirds (67 percent) of all applications created by developers completing the survey have embedded BI and analytics capabilities, the report says, and that's across "a wide spectrum of vertical industries" (health care, education, and utilities are the top users).
The report says the top three uses of embedded analytics are dashboards, self-service data analysis, and reports. However, its use in advanced analytics is worth watching. "Advanced and predictive analytics is making the biggest jump in popularity and will soon rival the big three in importance."
The report gives executives, product managers, and technology leaders an inside look at why and how organizations are embedding analytics capabilities in their applications. Embedded analytics isn't just a buzzword, or a "nice-to-have" feature that lies unused. The survey found that 43 percent of users actually employ embedded analytics on a regular basis -- and that number, the report predicts, will jump to 52 percent within two years.
Embedded analytics is also helping to expand overall user adoption and use of analytics applications. That 43 percent figure is even more impressive when you consider that traditional BI adoption typically tops out at 30 percent, says Logi Analytics (quoting Gartner).
Where does this analytics occur? Inline analytics -- which integrates with data, security, and the app's user interface -- was found in 53 percent of implementations. (The report notes that the popularity of inline analytics is "likely related to the fact that most third-party analytics applications can be embedded in this way.") Infused analytics -- where the analytics is part of the app's work flow -- was in 31 percent of implementations. However, year-over-year comparisons with previous surveys show that we'll likely be seeing deeper embedding over time. "Specifically, we see the adoption of both inline and infused analytics increasing, while standalone and gateway implementations are decreasing," the report says.
Of these two kinds of analytics, inline analytics is positioned to take the lead. "Within three years, it will be impossible to tell the difference between an application and the analytics content embedded within it. Just as B2C applications embed analytics as a natural part of the user experience, B2B applications will continue to move from bolt-on approaches to more infused implementations. Users will only have to use their preferred application, not two or more applications, in their daily work."
Embedded analytics isn't just good for end users; it's also good for vendor revenue. "The vast majority of commercial software and SaaS providers have increased revenue through embedded analytics, charging 24 percent on top of their core offerings." Analytics also accounts for "a relative value of 45 percent of the overall application." (Fifty-nine percent say embedded BI and analytics also lets them improve customer satisfaction and 55 percent say it's a product differentiator.)
If you're an applications provider, beware. Two-thirds of all application providers surveyed plan to invest in embedded analytics in the next year, so "Consider yourself warned: Even if analytics isn't in your plan, it's certainly in your competition's road map."
Over 500 business and technology professionals completed the survey, representing product management, product development, software engineering, and IT, as well as executives in commercial independent software vendors (ISVs), Software-as-a-Service (SaaS) providers, and non-commercial IT-managed applications used by internal staff and partners. Three-quarters of them were from North America and 23 percent from the UK. Just over a quarter (26 percent) were Logi Analytics customers.
We asked Brian Brinkmann, VP of product at Logi Analytics, about his reactions to the survey results.
"One result that jumped out to us was the uptick in adoption of advanced and predictive analytics," he said. "For many years, the industry has talked about advanced analytics as the next big thing. This is first time in the four years we've conducted this survey where advanced and predictive analytics are looking to rival the 'Big Three' of BI functionality: dashboards, reports, and self-service analysis.
"What's most encouraging is that by embedding advanced analytics in the applications people use every day, instead of just keeping the insights in the hands of data scientists, organizations can share those findings with a much broader group of individuals. These individuals can more quickly take action and offer feedback on the whether those advanced models are producing the expected results. The front-line workers are more effective and the data scientists can validate or invalidate their models far more quickly. This collaboration has the potential to offer noticeable performance improvement for the organization."
What the single most important fact or trend an enterprise business user should take away from this report?
According to Brinkmann , "When building a data-driven organization and culture, it is imperative that relevant data and information is available in the context of the applications and portals people access every day to drive adoption and use. Analytical tools have been around for years, but adoption has held steady for nearly a decade.
"It is clear that tools themselves are not solving the adoption issue. Organizations have consistently told us through our survey research and interviews that there is simply no easier way to increase analytics adoption than to embed insights into the applications and processes that run the business. Adoption rates of analytics embedded in applications is nearly double that of standalone tools."
The full report is available for download at http://www.logianalytics.com/report/state-embedded-analytics-2016/, though registration is required.
James E. Powell is the editorial director of TDWI, including the Business Intelligence Journal and Upside newsletter.