The same data has two shapes. Wide is one row per thing with a column per measurement; long is one row per measurement. Analysis is easier in one shape and output is easier in the other, and converting between them is a normal step, not a sign of trouble.
About 2 min of reading
Pivot long rows into a grid.
Long data
Before you run it, say out loud what it will print. Then open the answer.
Print the grid with a gap rule
Before you run it, say out loud what it will print. Then open the answer.
Before you write anything, put these lines back in the right order. Click a line to move it up one step. This support step is easier than writing code from scratch, and it is here to make the last example stick.
Read the first example one line at a time. A line that prints shows what appears on screen; the rest set things up quietly.
If your columns are jan, feb, mar, the month is a value hiding in a name. Any code that adds a month breaks, and any code that averages across columns averages the wrong thing. Melt those columns into a month field.
A grid with gaps has cells that do not exist. Decide whether absent means zero or unknown, and fill the grid consistently. The two answers give different totals, and the difference is usually the whole bug.
One row per observation: (region, month, value). It extends by adding rows, never by adding columns, so a new month needs no code change. This is the shape to store and the shape a database wants.
Months across the top, regions down the side, so a reader scans a grid. Build it from long at print time; never store it, because a new month means a new column and old code that does not know about it.
grid = {} then grid.setdefault(row['region'], {})[row['month']] = row['value']. That is the whole pivot. The reverse is a double loop over the grid emitting one row per cell.
Each one is defined here on this site. Open the entry for the full definition and the other words it points at.