Hi Katerina - while I don't have any examples to share, generally, the approach to creating turnover insights is often to answer the question, "How do the collaboration and network attributes of terminated employees (in the months leading to their exit) differ from their retained peers?" To do this, you'll need a couple of things:
- Accumulated collaboration data from employees that eventually leave. Once Viva Insights is processing data on a weekly basis, this will be gathered automatically from active mailboxes until the employee leaves the organization and the mailbox is deactivated. Note that when the product is first initialized, it may not have any data from terminated employees, unless the O365 admin has maintained and licensed their inactive mailboxes.
- A flag in the organizational (HR) data file that conveniently identifies terminated employees from active employees, as well as a column that includes their date of termination. Alternatively, you could assume that any licensed employee with zero collaboration activity is terminated, but including this in the uploaded descriptive organizational/HR data file makes the exercise much simpler.
With those elements in place, you can start building out turnover insights, either with simple comparative analysis (e.g., "Average collaboration hours for terminated engineers over the three months prior to their exit compared to average collaboration hours for retained engineers"), or using correlation or regression analysis.
