Monthly exports often arrive as separate files. Read them all with one map() and stack them with list_rbind(); give names_to = "month" and the list's names become a column, so every row says which file it came from.
In this lesson each export is a text string in the named vector exports, so read_csv(I(x)) reads it the way it would read a file. On your own machine you would map over list.files("exports", full.names = TRUE) instead.
list_rbind() matches columns by name, so a file whose columns come in another order still lines up. It stops when one file reads a column as text and another as a date or a number, which happens when a file is empty and read_csv() has nothing to guess from. Fix it by stating the types by name: col_types = cols(store = "c", date = "D", sales = "d").
Read every export with one function and stack them, keeping where each row came from.