with i as (
select
        client_id,
        DATE(MIN(case when action = 'visit' then hitdatetime else null end)) as fst_visit_dt,
        DATE(MIN(case when action = 'registration' then hitdatetime end)) as registration_dt,
        MAX(case when action = 'registration' then 1 else 0 end) as is_registration
from
        user_activity_log
group by
        client_id
    ),
    y as (
select
        client_id,
        SUM(payment_amount) as total_payment_amount
from
        user_payment_log upl
group by
        client_id
    )

select
    ua.client_id,
    ua.utm_campaign,
    i.fst_visit_dt,
    i.registration_dt,
    i.is_registration,
    y.total_payment_amount
from
    user_attributes ua
left join i on
    ua.client_id = i.client_id
left join y on
    y.client_id = ua.client_id
limit 10;