MySQL find Count grouped by per month - mysql

Question

For my problem the general structure of the tables is:
the Workers are located in different Branches (Branch table).
Prospective customer register (Registration table) as a Customer (Customer table)
and can order the products to buy (Order table).
Branch Table:
+------------+--------------+-----------------+
| 'branchId' | 'street' | 'city' |
+------------+--------------+-----------------+
| 'B002' | 'Clover Dr' | 'London' |
| 'B003' | 'Main St' | 'Glagsow' |
| 'B004' | 'Manse Rd' | 'Bristol' |
| 'B005' | 'Deer Rd' | 'London' |
| 'B007' | 'Argyll St' | 'Los Angeles' |
| 'B008' | 'Mission St' | 'San Francisco' |
| 'B009' | 'SOMA' | 'San Francisco' |
+------------+--------------+-----------------+
Customer Table:
+--------------+----------+-----------+-----------------+
| 'customerId' | 'fName' | 'lName' | 'telNo' |
+--------------+----------+-----------+-----------------+
| 'CR56' | 'Aline' | 'Stewart' | '0141-848-1825' |
| 'CR58' | 'Jacky' | 'Ho' | '0123-1325434' |
| 'CR62' | 'Mary' | 'Tregar' | '01224-196720' |
| 'CR74' | 'Mike' | 'Ritchie' | '01475-392178' |
| 'CR76' | 'John' | 'Kay' | '0207-774-5632' |
+--------------+----------+-----------+-----------------+
Registration Table:
+--------------+------------+------------+-----------------------+
| 'customerId' | 'branchId' | 'workerId' | 'joiningDate' |
+--------------+------------+------------+-----------------------+
| 'CR56' | 'B003' | 'SG37' | '2004-05-02 12:00:00' |
| 'CR58' | 'B003' | 'SA9' | '2004-05-03 12:00:00' |
| 'CR62' | 'B007' | 'SA9' | '2004-05-01 12:00:00' |
| 'CR74' | 'B004' | 'SG37' | '2004-04-04 12:00:00' |
| 'CR76' | 'B005' | 'SL41' | '2004-03-03 12:00:00' |
+--------------+------------+------------+-----------------------+
Order Table:
+--------------+---------------+-----------------------+
| 'customerId' | 'productId' | 'orderDate' |
+--------------+---------------+-----------------------+
| 'CR56' | 'PA14' | '2004-05-04 11:30:00' |
| 'CR62' | 'PA14' | '2004-05-04 14:00:00' |
| 'CR56' | 'PG36' | '2004-06-07 11:00:00' |
| 'CR56' | 'PG4' | '2004-04-14 12:05:00' |
| 'CR76' | 'PG4' | '2004-04-04 10:15:00' |
+--------------+---------------+-----------------------+
I am trying to form a query to find the number of orders per Branch within 1, 2, and 3 months of client Registration.
Let's say for example
+----------+------------+-----------------+
| 'months' | 'branchId' | 'numberOfOrder' |
+----------+------------+-----------------+
| 1 | 'B003' | 2 |
| 2 | 'B004' | 1 |
+----------+------------+-----------------+
I tried to group the table by month and date but I am stuck and not able to proceed forward.
Does anyone has any ideas and help me unblock?
I started doing something like this, but I am completely lost at the moment.
SELECT
COUNT(DISTINCT o.orderDate) AS 'count'
FROM
Order o, Registration r
WHERE
o.orderDate BETWEEN DATE('2001-01-01') AND DATE('2005-01-31')
GROUP BY YEAR(o.orderDate), MONTH(o.orderDate);
But this seems I am pretty far from what I am trying to achieve.

Answer 1

I'm not entirely sure of what your desired result is, but with this query you can get the count of orders, per branch, within 3 months after registration.
SELECT
reg.branchId,
COUNT(reg.branchId) AS 'orderCount'
FROM `order` AS ord INNER JOIN `registration` AS reg
ON ord.customerId = reg.customerId
WHERE reg.joiningDate BETWEEN reg.joiningDate AND DATE_ADD(reg.joiningDate, INTERVAL 3 MONTH)
GROUP BY reg.branchId
Result
Is this what you wanted to do?

Answer 2

Your desired result has nothing in common with your data.
So i assume you want the Order count for every branch.
I added the year also, because it is usually needed and doesn't bother if your data don't go over one year
Update:
Now ot only select orders which was place in the 3 month since the a customer joined
.It is limited by the date_add in the where clause
CREATE TABLE registration
(`customerId` varchar(4), `branchId` varchar(4), `workerId` varchar(4), `joiningDate` datetime)
;
INSERT INTO registration
(`customerId`, `branchId`, `workerId`, `joiningDate`)
VALUES
('CR56', 'B003', 'SG37', '2004-05-02 12:00:00'),
('CR58', 'B003', 'SA9', '2004-05-03 12:00:00'),
('CR62', 'B007', 'SA9', '2004-05-01 12:00:00'),
('CR74', 'B004', 'SG37', '2004-04-04 12:00:00'),
('CR76', 'B005', 'SL41', '2004-03-03 12:00:00')
;
✓
✓
CREATE TABLE `order`
(`customerId` varchar(4), `productId` varchar(4), `orderDate` datetime)
;
INSERT INTO `order`
(`customerId`, `productId`, `orderDate`)
VALUES
('CR56', 'PA14', '2004-05-04 11:30:00'),
('CR62', 'PA14', '2004-05-04 14:00:00'),
('CR56', 'PG36', '2004-06-07 11:00:00'),
('CR56', 'PG4', '2004-04-14 12:05:00'),
('CR76', 'PG4', '2004-04-04 10:15:00')
;
✓
✓
SELECT MONTH(o.`orderDate`),r.branchId, COUNT(*) numberOfOrder
FROM registration r inner join `order` o ON r.`customerId` = o.`customerId`
WHERE o.`orderDate` BETWEEN r.`joiningDate` AND DATE_ADD(r.`joiningDate`, INTERVAL 3 MONTH)
GROUP BY YEAR(o.`orderDate`),MONTH(o.`orderDate`),r.branchId
MONTH(o.`orderDate`) | branchId | numberOfOrder
-------------------: | :------- | ------------:
4 | B005 | 1
5 | B003 | 1
5 | B007 | 1
6 | B003 | 1
db<>fiddle here

Answer 3

You may try below query i guess, having moths calculated on the basis of differences in orderDate and joiningDate -
SELECT abs(ceil(datediff(o.`orderDate`, r.`joiningDate`)/30)) months_join,r.branchId, COUNT(*) numberOfOrder
FROM registration r inner join `order` o ON r.`customerId` = o.`customerId`
GROUP BY YEAR(o.`orderDate`), abs(ceil(datediff(o.`orderDate`, r.`joiningDate`)/30)),r.branchId