GroupBy + Pivot Tables in Pandas

Turn raw data into clear, Excel-like summary tables using GroupBy and pivot_table()

If you've started using groupby(), you're already doing powerful aggregations. But sometimes you need your summary to look like a proper table — with rows, columns, and values in the right places — just like pivot tables in Excel.

That's exactly what df.pivot_table() does. It uses GroupBy under the hood but automatically reshapes the result into a clean, readable table.

Our Sample DataFrame

Let's use sales data from a small company:

    
        
            
            Region
            Product
            Year
            Sales
        
    
    
        
            0
            North
            Laptop
            2023
            1200
        
        
            1
            North
            Phone
            2023
            800
        
        
            2
            South
            Laptop
            2023
            1100
        
        
            3
            South
            Phone
            2023
            900
        
        
            4
            East
            Laptop
            2024
            1300
        
        
            5
            East
            Phone
            2024
            850
        
        
            6
            West
            Laptop
            2024
            1400
        
        
            7
            West
            Phone
            2024
            950
        
        
            8
            North
            Phone
            2024
            820
        
        
            9
            South
            Laptop
            2024
            1150
        
    


Simple GroupBy Recap

First, a quick GroupBy example (total sales by Region):

df.groupby('Region')['Sales'].sum()

Output (Series):


    

        

            
Region
            
Sales
        
    
    

        

            
East
            
2150
        
        

            
North
            
2820
        
        

            
South
            
3150
        
        

            
West
            
2350
        
    


Enter pivot_table() – GroupBy Made Beautiful

Use pivot_table() when you want:


    
Rows = one category (e.g., Region)
    
Columns = another category (e.g., Product)
    
Values = aggregated numbers (e.g., total Sales)


    
        
            Region
            Laptop
            Phone
        
    
    
        
            East
            1300
            850
        
        
            North
            1200
            1620
        
        
            South
            2250
            900
        
        
            West
            1400
            950
        
    


Add Totals and Multiple Aggregations

    
        
            Region
            Laptop
            Phone
            Total
        
    
    
        
            East
            1300
            850
            2150
        
        
            North
            1200
            1620
            2820
        
        
            South
            2250
            900
            3150
        
        
            West
            1400
            950
            2350
        
        
            Total
            6150
            4320
            10470
        
    


Key Parameters of pivot_table()


    
values — column to aggregate
    
index — rows (can be list for multi-level)
    
columns — columns (can be list)
    
aggfunc — 'sum', 'mean', 'count', etc. (or list for multiple)
    
margins=True — add row/column totals
    
fill_value=0 — replace NaN with 0


When to Use Which?


    
Use plain groupby() for quick summaries or when you don't need reshaping.
    
Use pivot_table() for reports, dashboards, or when you want an Excel-style cross-tab view.



Happy Learning, Ritesh! 🐼