Many organizations continue to perceive and use embedded BI highly. However, reported current use of embedded BI declined slowly from its high of 53 percent in 2017. Its perceived importance sits below the all-time high level reported in 2018. Despite these declines, 12-month adoption plans are steady and 24-month adoption plans increased. Possible explanations for these data include higher expectations of what constitutes embedded BI (vs. previous years), as well as the impact of COVID-19 pausing and resetting many deployments.
Embedded BI often represents one component of successful embedded BI implementations because the breadth of its potential uses and its ease of deployment helps bring data and analytics closer to associates in the enterprise applications and portals they use in their everyday work. Embedded BI also can enable an ability to do light analyses on data in dashboards and reports, without the user needing to copy and manipulate source data (often using spreadsheets). Although limited in comparison to more standard BI tools, capabilities from embedded BI also tend to be much more intuitive, requiring fewer resources to develop skills and train users.
Embedded BI tends to arrive organically and dynamically—usually indirectly as enhancements to enterprise applications and portals (from the providers), as well as to reports and dashboards. Often enough, these “bottom-up” enhancements also tend to arrive somewhat stealthily, thereby lacking immediate alignment with business goals, programs, and projects. Although less common at this point, planned embedded BI deployments by organizations provide a means by which to address targeted-user analytical needs more tactically through more overlay capabilities, without requiring extensive application-upgrade investments and retraining.
Declining current usage, slipping importance, but expected higher future use present an interesting set of challenges that could make data leaders more complacent toward embedded BI just at the time its dynamics are changing and require more, not less, of their attention.
Because its ease of deployment—including more stealthy unplanned instances through extensions to applications and improvements to portals—embedded BI often can conceal the lack of a coordinated strategy that aligns to an overall data and analytics program and its value to the business. This can create a strong risk of an execution gap and, possibly, the need for better-coordinated implementations.
A successful implementation requires planning, alignment to strategy, monitoring, and maintenance beyond an initial rollout. A significant portion of embedded BI may come not from planned deployments but through extensions and enhancements to enterprise applications and portals.
Furthermore, data leaders need to consider and manage embedded BI as one of many channels and capabilities the organization uses as part of its overall data and analytics strategy—to help enable and reinforce information democracy and empower users to make more-frequent, better-informed data-driven business decisions.
Embedded BI implementations done in this way will enable organizations to take more opportunistic views of daily activities, and provide opportunities for better visibility, depth, and timeliness to enterprise and third-party data.