You are going to build a dashboard.
You’ve got the question you need to answer…check!
And you have the data to build the required calculations…check check!
But if you don’t walk that last mile with your target end users about *how* they want to see and interact with the data in your BI tool, you may waste hours of time and piles of money building the wrong thing in the wrong way.
How do I know this? Been there, done that. Experience is a great teacher.
For example, if someone wants to see a count of something like “Orders by Order Date”, that’s easy enough to build and show on a line chart. If someone wants to see a count of those same Orders, but now “Orders by Ship Date”, that’s also pretty straightforward to build and show on a line chart. But when they ask to see the same Orders counted once by “Order Date” and again by “Ship Date” together on the same date axis of a line chart, that might be a problem in your BI tool. With Power BI, you’d need to change the underlying data model and revise your calculations to get this right; not a showstopper, but not a trivial thing, either.
Or if someone assumes your BI tool can do everything Excel can, and uses Excel as their frame of reference for “baseline features every BI tool must already have”, you will be in trouble if you don’t show them how they can interact with your BI tool before building and delivering a dashboard. This *could* be a showstopper if a reasonable workaround wasn’t possible, or at least a major fail on “first impressions”. Either way, time and money are lost and confidence is shaken.
Before you spend a lot of time building data models and dashboards, you need to ask/understand how your target audience wants to interact with them and tell/communicate how they can interact with them. Ask about it. Poke at it. Mock something up. Create some sample data to show how different types of visualizations will look and behave. Show what types of user interaction with visualizations are allowed, and disallowed, by the features and functions of your BI tool.
And stay within the boundaries of the foundational principles I describe in my previous blog post “Tool Agnostic Analytics: Building Effective Dashboards”.
How your target audience wants to see and interact with a dashboard can and will affect how you build the underlying data model and calculations. Walk that last mile with them before you do.