I expected some kind of more obvious relationship between restaurants and groceries. This is absolute $ spent on each in a given month, as a scatter plot. This is for a family of four living in San Mateo in the Bay Area. Are your results any different?

Pro tip: you can export all your transactions from mint, making it a unified way to get all your banking information. I used to do my own manual categorization. The categorization at Mint is junk because it isn’t 100% and the interface isn’t good for editing like a spreadsheet it. So now the process is mint export CSV -> python script -> csv -> google spreadsheets. I should probably just skip the last step, but it is a good way to do manual editing which seems to always be necessary. For example, checks that I didn’t label in Bank of America’s interface (for non trivial amounts to preschools) show up as uncategorized if the amount varies from the recurring amount.

Side note: I wish google spreadsheets could do a directed scatter plot with a line connecting the dots chronologically. Alternatively, labels on each point would be nice.