your answer but not working
WITH daily_sums_table AS
(select city.name,
 daily_sales.day,
 salesman.id as salesman_id,
 sum(amount_earned) as sum_amount_earned
from daily_sales
join salesman on daily_sales.salesman_id = salesman.id
join city on salesman.city_id = city.id
group by 1,2,3),

average_daily_table AS
(select avg(sum_amount_earned) as avg_daily_sums_earned
from daily_sums_table)

select city.id,
city.name,
count(daily_sums_table.day)
--count(case when daily_sums_earned > avg_daily_sums_earned then 1 else 0 end)
from city
join salesman
on city.id = salesman.city_id
join daily_sales
on salesman.id = daily_sales.salesman_id
join daily_sums_table
on salesman.id = daily_sums_table.salesman_id
where sum_amount_earned > (select avg_daily_sums_earned from average_daily_table)
group by 1,2


------------------------------------------------------------------------------------------------------------------------
answer:
with daily_sums as (
select daily_sales.day,
city.id,
city.name,
sum(amount_earned)as sum_amount_earned
from daily_sales
join salesman on daily_sales.salesman_id = salesman.id
join city on salesman.city_id = city.id
group by daily_sales.day,city.id,city.name),

average_daily as
(select avg(sum_amount_earned) as avg_sum_amount_earned from daily_sums)

select daily_sums.id,
daily_sums.name,
count(daily_sums.day) from
daily_sums
where sum_amount_earned > (select avg_sum_amount_earned from average_daily)
group by daily_sums.id, daily_sums.name