Reputation: 23722
I have a table of sales leads:
CREATE TABLE "lead" (
"id" serial NOT NULL PRIMARY KEY,
"marketer" varchar(500) NOT NULL,
"date_set" varchar(500) NOT NULL
)
;
INSERT INTO lead VALUES (1, 'Joe', '05/01/13');
INSERT INTO lead VALUES (2, 'Joe', '05/02/13');
INSERT INTO lead VALUES (3, 'Joe', '05/03/13');
INSERT INTO lead VALUES (4, 'Sally', '05/03/13');
INSERT INTO lead VALUES (5, 'Sally', '05/03/13');
INSERT INTO lead VALUES (6, 'Andrew', '05/04/13');
I want to produce a report that summarizes the number of records each marketer has for each day. It should look like this:
| MARKETER | 05/01/13 | 05/02/13 | 05/03/13 | 05/04/13 |
--------------------------------------------------------
| Joe | 1 | 1 | 1 | 0 |
| Sally | 0 | 0 | 2 | 1 |
| Andrew | 0 | 0 | 0 | 1 |
What's the SQL query to produce this?
I have this example set up on SQL Fiddle: http://sqlfiddle.com/#!12/eb27a/1
Upvotes: 1
Views: 2263
Reputation: 29458
You can do it like this:
select
marketer,
count(case when date_set = '05/01/13' then 1 else null end) as "05/01/13",
count(case when date_set = '05/02/13' then 1 else null end) as "05/02/13",
count(case when date_set = '05/03/13' then 1 else null end) as "05/03/13",
count(case when date_set = '05/04/13' then 1 else null end) as "05/04/13"
from lead
group by marketer
Upvotes: 1
Reputation: 513
Pure SQL cannot produce such structure (it is two dimensional, but sql return plain list of records).
You could make query like this:
select marketer, date_set, count(id)
from lead
group by marketer, date_set;
And vizualise this data by your reporting system.
Upvotes: 1