Insights are provided through interactive data visualisations, such as charts, diagrams, filters, gauges, maps and tables often in combination as dashboards embedded within the system.
Here, though, the word "business" in "business-to-business software as a service", could also refer to organisational, operational use cases that ultimately benefit consumers (such as healthcare, for instance), e.g.: clinics & hospitals, care & correctional facilities, educational establishments (on/offline), government departments, municipalities, museums, not-for-profit organisations, overseers & regulators amongst others.
This is in contrast to traditional BI, which expects users to leave their workflow applications to look at data insights in a separate set of tools.
The objective would be intuitive, contextual analytics, consumed as regular web content, immersed into operational user experiences and workflows usable without any special knowledge or training required.
Once a root-cause has been investigated, the embedded analytics can place the user in the part of the host application to act, potentially at scale.
So rather than reschedule, reorder, reassign one thing, the embedded analytics can apply business rules and pass parameters to the host application to act 100 or 1,000 times, instead of once, where each action may be individually customised.