I have data for 3 years (2016, 2017, 2018) and I need to show how consumers moved from one banking product (savings, checking, both) to the next. Each row is a bank branch and the count are their customers. For example, I need to show of the 97K that had savings in 2016, 76K kept savings, 500 moved to checking, 2,700 moved to both, 17K dropped out, and 16K were totally new to savings.
I'd like to use a Sankey, but all of the examples I've looked at use a single metric with multiple dimensions. In my case I can either have one metric (members) and one dimension (Product) or multiple of each (a column for each year). I have the ability to pivot the data if need be.
Any thoughts?
I've attached a workbook with sample data.
Thanks!
-Wesley
In case anyone is interested in the solution, I ended-up creating a Sankey while leveraging a wider table. Each year was given a field for Product and Consumer Count. From there I leveraged this youtube video on Sankey diagrams (SANKEY DIAGRAM TABLEAU - YouTube). I created two Sankey diagrams, one for 2016-2017 and one for 2017-2018. Each diagram used the metrics from the latter year and the Product type from both years. This is how I was ultimately able to have one metric (ex 2017 Consumer Count) and two dimensions (2016 Product, 2017 Product).
This approach meant that I lost consumers that dropped between years, but captured new consumers and all of the changes.
