Question

SQL appears to be more complex than I anticipated. My problem: for each customer, I would like to show the Customer ID and the total number of orders placed in 2011.

My table looks like this

Table: Order_t
Order_ID    Order_Date    Customer_ID
--------    ----------    -----------
1001        10/21/2011    1
1002        10/25/2011    8
1003        10/26/2011    15
1004        10/27/2011    5
1005        11/24/2011    3
1006        11/27/2011    2
1007        11/28/2011    11
1008        12/3/2011     12
1009        12/5/2011     1
1010         1/16/2012    4

I would like my query to display a table like this:

Customer_ID    Orders_Placed
-----------    -------------
1              2
2              1
3              1
5              1
8              1
11             1
12             1
15             1

My current query is this (I am currently completely neglecting the Date part because I haven’t even figured out the grouping yet:

SELECT Customer_ID, SUM(Order_ID) AS Orders_Placed
FROM Order_t
GROUP BY Order_ID, Customer_ID

And this is my obviously wrong query:

Customer_ID    Orders_Placed
-----------    -------------
1              1001
8              1002
15             1003
5              1004
3              1005
2              1006
11             1007
12             1008
1              1009
4              1010

Thanks for help, but I would also like to understand where the problem is in my logic. What crucial part do I seem to not understand?

Answer 1

The problem with your logic is this

GROUP BY Order_ID, Customer_ID

Which means each combination of (Order_ID, Customer_ID) is put in a different GROUP. Since Order_ID alone is unique, practically no grouping is happening.

To do it correctly, you need to GROUP BY the Customer_ID (it reads like what you need, doesn’t it), then COUNT the Orders. Finally, add the date filter as well.

SELECT Customer_ID, COUNT(Order_ID) AS Orders_Placed
FROM Order_t
WHERE Order_Date >= #1/1/2011# and Order_Date < #1/1/2012#
GROUP BY Customer_ID