I got (ERROR: more than one row returned by a subquery used as an expression) or incorrect count/sum inside subquery - sql

Question

I would like to make a query that will show me the total count and sum of sales, and for 2 workers in the same rows.
When I'm not using group by in a subquery I get a count for total sales for this worker not counted for each month. But when I try to use Group By clause in a subquery it shows me this error:
PG::CardinalityViolation:
ERROR: more than one row returned by a subquery used as an expression
And that makes sense but how I can count and sum amount of sales per month for each worker? Is there any possibility of transferring the value of Group By clause from query to a subquery? Should I be using subquery for this specific issue or maybe I can somehow make it easier?
SELECT
DATE_PART('month', payment_date) as month,
COUNT(payment_id) AS total_count,
SUM(amount) AS total_amount,
(SELECT COUNT(payment_id)
FROM payment
WHERE staff_id = 1
GROUP BY DATE_PART('month', payment_date)) AS mike_count,
(SELECT SUM(amount)
FROM payment
WHERE staff_id = 1
GROUP BY DATE_PART('month', payment_date)) AS mike_amount,
(SELECT COUNT(payment_id)
FROM payment
WHERE staff_id = 2
GROUP BY DATE_PART('month', payment_date)) AS jon_count,
(SELECT SUM(amount)
FROM payment
WHERE staff_id = 2
GROUP BY DATE_PART('month', payment_date)) AS jon_amount
FROM
payment
GROUP BY
DATE_PART('month', payment_date)
ORDER BY
DATE_PART('month', payment_date);
This is the outcome w/o the group by clause in subqueries:
month |total_count |total_amount |mike_count |mike_amount |jon_count |jon_amount
2 |2016 |0.835184E4 |7292 |0.3025212E5 |7304 |0.3105992E5
3 |5644 |0.2388656E5 |7292 |0.3025212E5 |7304 |0.3105992E5
4 |6754 |0.2855946E5 |7292 |0.3025212E5 |7304 |0.3105992E5
5 |182 |0.51418E3 |7292 |0.3025212E5 |7304 |0.3105992E5

Answer 1

Use conditional aggregation instead:
SELECT DATE_PART('month', payment_date) as month,
count(payment_id) AS total_count,
SUM(amount) AS total_amount,
SUM(CASE WHEN staff_id = 1 THEN 1 ELSE 0 END) as mike_count,
SUM(CASE WHEN staff_id = 1 THEN amount ELSE 0 END) as mike_amount,
SUM(CASE WHEN staff_id = 2 THEN 1 ELSE 0 END) as jon_count,
SUM(CASE WHEN staff_id = 2 THEN amount ELSE 0 END) as jon_amount
FROM payment
GROUP BY DATE_PART('month', payment_date)
ORDER BY DATE_PART('month', payment_date);
DATE_PART() suggests that you are using Postgres. If so, I would recommend the FILTER clause:
SELECT DATE_PART('month', payment_date) as month,
COUNT(payment_id) AS total_count,
SUM(amount) AS total_amount,
COUNT(*) FILTER (WHERE staff_id = 1) as mike_count,
SUM(amount) FILTER (WHERE staff_id = 1) as mike_amount,
COUNT(*) FILTER (WHERE staff_id = 2) as jon_count,
SUM(amount) FILTER (WHERE staff_id = 2) as jon_amount
FROM payment
GROUP BY DATE_PART('month', payment_date)
ORDER BY DATE_PART('month', payment_date);