When using the NTILE() function in SQL Server, we must include an OVER clause clause with an ORDER BY clause.

If you’re getting error msg 4112 that reads “The function ‘NTILE’ must have an OVER clause with ORDER BY” when using the NTILE() function, it’s because, although you’re (correctly) including an OVER clause, you’re omitting the ORDER BY clause.

To fix this error, add an ORDER BY clause to the OVER clause.

Example of Error

Here’s an example of code that produces the error:

SELECT
    VendorId,
    ProductName,
    ProductPrice,
    NTILE( 3 ) OVER ( 
        PARTITION BY VendorId
        ) AS NTILE
FROM Products;

Result:

Msg 4112, Level 15, State 1, Line 5
The function 'NTILE' must have an OVER clause with ORDER BY.

This error occurred because the OVER clause doesn’t have an ORDER BY clause.

We’ll still get the error if we put an ORDER BY clause at the end of the query, but not in the OVER clause:

SELECT
    VendorId,
    ProductName,
    ProductPrice,
    NTILE( 3 ) OVER ( 
        PARTITION BY VendorId
        ) AS NTILE
FROM Products
ORDER BY ProductPrice;

Result:

Msg 4112, Level 15, State 1, Line 5
The function 'NTILE' must have an OVER clause with ORDER BY.

So the NTILE() function must have an OVER clause with its own ORDER BY clause.

Solution

To fix this issue, simply add an ORDER BY clause to the OVER clause:

SELECT
    VendorId,
    ProductName,
    ProductPrice,
    NTILE( 3 ) OVER ( 
        PARTITION BY VendorId
        ORDER BY ProductPrice
        ) AS NTILE
FROM Products;

Result:

VendorId  ProductName                      ProductPrice  NTILE
--------  -------------------------------  ------------  -----
1001      Long Weight (green)              11.99         1    
1001      Long Weight (blue)               14.75         1    
1001      Left handed screwdriver          25.99         2    
1001      Right handed screwdriver         25.99         3    
1002      Sledge Hammer                    33.49         1    
1003      Hammock                          10            1    
1003      Straw Dog Box                    55.99         2    
1003      Chainsaw                         245           3    
1004      Bottomless Coffee Mugs (4 Pack)  9.99          1    
1004      Tea Pot                          12.45         2    

This time the function worked as expected, without error.

In this example I also used the PARTITION BY clause to partition the result set, but this is optional. We can omit the PARTITION BY clause but we can’t omit the ORDER BY clause.