pivot_wider() is the reverse: it spreads one column's values into columns. ticket_months |> pivot_wider(names_from = month, values_from = tickets) gives one row per team and one column per month, the layout a report or a spreadsheet wants.
Combinations that don't exist in the long table become NA. Decide what that means. If Reports logged no tickets in February, the honest number is 0: values_fill = 0. If the data is simply missing, leave it NA.
New columns appear in the order the values first appear in the long table. Keep long data for analysis and charts, and widen at the end, for the people who will read the table.
One row per team, one column per month, and 0 where no tickets were logged.