Well, it’s kind of a tricky problem. Every time we’ve considered circular charts we’ve come to the conclusion that they weren’t really the best solution for the problem we were trying to solve.
Pie charts (for our use case) have two big problems:
Spending is usually really skewed towards rent and bills so any pie chart of spending is likely to have maybe a 50% of the chart precision gone in just one category. That leaves very little room for the rest of spending.
Categories with little spending take so very little space on screen (because they are proportional to the number) that are really difficult to interact with (tap to dive in or select, etc.) because of the lack of tappable surface (a classic problem of Fitt’s law).
So, we won’t probably have pie charts until we want to produce some kind of reports where we don’t want you to interact with or that only compares a few categories or merchants (for example: Uber vs TfL cost).
The bottom line is we never design data visualisations from top (“oh yes, let’s make a nice graph here… what can we populate it with?”). We take the data and the story that we want to tell and once we understand it we try to find the best possible visualisation for it. Until now the answer has never been I’m afraid… but we’ll keep working