If truth be told, dashboards were one of the first big successes and then, the first big failures, of people analytics.
They were a success in that they were a shiny way to efficiently get data out to managers.
But they were a failure in that after an initial surge of interest, managers stopped looking at them.
Today we have reached a stage where we have a much more refined understanding of what dashboards are for.
Here are the five main reasons for using dashboards:
1) Raising questions
My favorite insight about dashboards comes from an analytics professional who said that dashboards exist to raise questions, not to answer them. This sets up a realistic expectation of what dashboards can achieve, and it raises the need for analytics professionals who can collaborate with business leaders to dig into the quantitative and qualitative evidence to help explain any surprising information surfaced by the dashboard.
2) Providing day-to-day operational insights
There are some dashboards that managers don’t get tired of looking at because they guide day-to-day operations. While this is somewhat rare in HR, it still shows up in high-volume work such as recruiting and managing the hourly workforce. Talent acquisition professionals will want to track how many requisitions are open, which have been open for too long, and how the process for crucial hires is moving along. Managers of hourly positions may regularly track data on overtime and scheduling. Operational dashboards are the best-loved type of dashboards.
3) Data that managers don’t care about…until they do
One uninspiring but useful role for dashboards is to provide managers with easy access to data that they rarely want but when they do, they expect it to be at hand. For example, managers may suddenly have the desire to know the number of part-timers in their department or the number of positions they budgeted for, but have remained unfilled. Having such data available in a dashboard, even if it is rarely accessed, is important for HR’s credibility.
4) Data relevant for periodic (mainly annual) business processes
There are some pieces of data that managers will care about at specific times of the year to help them with business processes. For example, they will want lots of information relevant to performance for performance reviews and may want to see engagement scores and diversity metrics once a year. Dashboards should separate metrics that are of interest only at specific times of the year from the ones providing day-to-day operational insights.
5) For giving data that managers want now, but won’t want in the future
One seasoned analytics professional suggested the metrics in a dashboard should change by about 30% every year. For example, managing interns may be a big concern this year and managers want lots of data about this group now, but next year another issue will have arisen, and the intern data will be of no interest. Analytics professionals should be ready to remove metrics that have lost their relevance.
We’ve learned that dashboards with lots of cool visualization features may get a good reception, but not have any enduring value.
We’ve also learned that asking managers what data they want is rarely a reliable way of discovering what information they will use and when.
People analytic professionals need a well-informed idea of how any piece of data will be used and design their dashboards accordingly.
The checklist of five types of dashboards given above can be a handy reminder of how dashboards tend to be used.