addressed Pablo's comments, removed the repetitive casts, added some nut_null tests and fixed some of the names and descriptions discrepancies

models/intermediate/edeposit/int_edeposit__agg_fee_per_user.sql

@@ -11,15 +11,15 @@ with
         select
             id_booking,
             is_cancelled,
+            id_user_partner,
             channel,
-            cast(checkin_at_utc as date) as checkin_date_utc,
+            checkin_date_utc,
-            cast(checkout_at_utc as date) as checkout_date_utc,
+            checkout_date_utc,
             creation_at_utc,
             verification_status,
-            id_user_host,
             case
                 when verification_status in {{ ok_status }}
-                then cast(nightly_fee_local as float) * number_nights
+                then nightly_fee_local * number_nights
                 else 0
             end as ok_status_fee,
             case
@@ -30,33 +30,50 @@ with
             to_char(creation_at_utc, 'YYYY-MM') as year_month_created,
             to_char(checkout_at_utc, 'YYYY-MM') as year_month_checkout
         from int_edeposit__verifications
-        where version = 'V2'  -- This version V2 indicates records for e-deposit
+        where version = 'V2'
+    ),
+    calculated_metrics as (
+        select
+            id_user_partner,
+            year_month_created,
+            year_month_checkout,
+            count(distinct id_booking) as bookings_per_month,
+            sum(cast(is_cancelled as integer)) as cancelled_per_month,
+            sum(cast(is_cancelled as integer))::decimal
+            / count(distinct id_booking) as cancelled_ratio,
+            sum(ok_status_fee) as sum_ok_status_fee,
+            sum(rejected_fee) as sum_rejected_fee,
+            case
+                when
+                    sum(cast(is_cancelled as integer))::decimal
+                    / count(distinct id_booking)
+                    >= {{ cancellation_threshold }}
+                then true
+                else false
+            end as is_cancellation_threshold_surpassed
+        from edeposit_records
+        group by year_month_created, year_month_checkout, id_user_partner
     )
 select
+    id_user_partner,
     year_month_created,
     year_month_checkout,
-    id_user_host,
+    bookings_per_month,
-    count(distinct id_booking) as bookings_per_month,
+    cancelled_per_month,
-    sum(cast(is_cancelled as integer)) as cancelled_per_month,
+    cancelled_ratio,
-    sum(cast(is_cancelled as integer))::decimal
+    is_cancellation_threshold_surpassed,
-    / count(distinct id_booking) as cancelled_ratio,
     case
-        when
+        when is_cancellation_threshold_surpassed
-            sum(cast(is_cancelled as integer))::decimal / count(distinct id_booking)
+        then cancelled_per_month * {{ cancellation_fee }}
-            >= {{ cancellation_threshold }}
-        then sum(cast(is_cancelled as integer)) * {{ cancellation_fee }}
         else 0
     end as sum_cancelled_fee,
-    sum(ok_status_fee) as sum_ok_status_fee,
+    sum_ok_status_fee,
-    sum(rejected_fee) as sum_rejected_fee,
+    sum_rejected_fee,
-    case
+    sum_ok_status_fee
-        when
+    + sum_rejected_fee
-            sum(cast(is_cancelled as integer))::decimal / count(distinct id_booking)
+    + case
-            >= {{ cancellation_threshold }}
+        when cancelled_ratio >= {{ cancellation_threshold }}
-        then sum(cast(is_cancelled as integer)) * {{ cancellation_fee }}
+        then cancelled_per_month * {{ cancellation_fee }}
         else 0
-    end
+    end as total_revenue
-    + sum(ok_status_fee)
+from calculated_metrics
-    + sum(rejected_fee) as total_revenue
-from edeposit_records
-group by year_month_created, year_month_checkout, id_user_host

models/intermediate/edeposit/int_edeposit__verifications.sql

@@ -5,9 +5,9 @@ with
 select
     -- note that these ids are not the same as the ones found in Core DWH
     -- they are completely unrelated
-    id as id_verification,
+    id_verification,
     id_booking,
-    id_user as id_user_host,
+    id_user as id_user_partner,
     id_listing as id_accommodation,
     version,  -- V1 for Guesty and V2 for E-deposit
     case
@@ -15,14 +15,16 @@ select
     end as verification_source,
     verification_status,
     verification_status_reason,
-    nightly_fee_local,
+    cast(nightly_fee_local as float),
     cast(checkout_at_utc as date) - cast(checkin_at_utc as date) as number_nights,
     email_flag,
     phone_flag,
     watch_list,
     channel,
     checkin_at_utc,
+    cast(checkin_at_utc as date) as checkin_date_utc,
     checkout_at_utc,
+    cast(checkout_at_utc as date) as checkout_date_utc,
     is_cancelled,
     cancelled_at_utc,
     user_email,

models/intermediate/edeposit/schema.yaml

@@ -19,9 +19,11 @@ models:
         data_type: text
         description: "unique Superhog generated id for a booking"
 
-      - name: id_user_host
+      - name: id_user_partner
         data_type: text
-        description: "unique Superhog generated id for host"
+        description: "unique Superhog generated id for partner"
+        tests:
+          - not_null
 
       - name: id_accommodation
         data_type: text
@@ -56,7 +58,7 @@ models:
         description: "short explanation for status"
 
       - name: nightly_fee_local
-        data_type: text
+        data_type: double precision
         description: "fee charged per night"
 
       - name: number_nights
@@ -83,10 +85,18 @@ models:
         data_type: timestamp without time zone
         description: "Timestamp of checkin for the booking"
 
+      - name: checkin_date_utc
+        data_type: date
+        description: "Timestamp of checkin for the booking"
+
       - name: checkout_at_utc
         data_type: timestamp without time zone
         description: "Timestamp of checkout for the booking"
 
+      - name: checkout_date_utc
+        data_type: date
+        description: "Timestamp of checkout for the booking"
+
       - name: is_cancelled
         data_type: boolean
         description: ""
@@ -184,19 +194,19 @@ models:
           combination_of_columns:
             - year_month_created
             - year_month_checkout
-            - id_user_host
+            - id_user_partner
     columns:
+      - name: id_user_partner
+        data_type: text
+        description: "unique id value for user partner"
+
       - name: year_month_created
         data_type: text
-        description: "first day of month of created date"
+        description: "year and month of created date"
 
       - name: year_month_checkout
         data_type: text
-        description: "first day of month of check-out date"
+        description: "year and month of check-out date"
-
-      - name: id_user_host
-        data_type: text
-        description: "unique id value for user host"
 
       - name: bookings_per_month
         data_type: bigint
@@ -210,6 +220,12 @@ models:
         data_type: numeric
         description: "ratio of cancelled bookings over total bookings"
 
+      - name: is_cancellation_threshold_surpassed
+        data_type: boolean
+        description: "true if the cancellation ratio is higher than the set threshold"
+        tests:
+          - not_null
+
       - name: sum_cancelled_fee
         data_type: numeric
         description: "sum of fees charged for cancelled bookings

models/reporting/edeposit/edeposit__agg_fee_per_user.sql

@@ -5,10 +5,11 @@ with
 select
     year_month_created as year_month_created,
     year_month_checkout as year_month_checkout,
-    id_user_host as id_user_host,
+    id_user_partner as id_user_partner,
     bookings_per_month as bookings_per_month,
     cancelled_per_month as cancelled_per_month,
     cancelled_ratio as cancelled_ratio,
+    is_cancellation_threshold_surpassed as is_cancellation_threshold_surpassed,
     sum_cancelled_fee as sum_cancelled_fee,
     sum_ok_status_fee as sum_ok_status_fee,
     sum_rejected_fee as sum_rejected_fee,

models/reporting/edeposit/schema.yaml

@@ -14,19 +14,25 @@ models:
           combination_of_columns:
             - year_month_created
             - year_month_checkout
-            - id_user_host
+            - id_user_partner
     columns:
+      - name: id_user_partner
+        data_type: text
+        description: "unique id value for user partner"
+        test:
+          - not_null
+
       - name: year_month_created
         data_type: text
-        description: "first day of month of created date"
+        description: "year and month of created date"
+        test:
+          - not_null
 
       - name: year_month_checkout
         data_type: text
-        description: "first day of month of check-out date"
+        description: "year and month of check-out date"
-
+        test:
-      - name: id_user_host
+          - not_null
-        data_type: text
-        description: "unique id value for user host"
 
       - name: bookings_per_month
         data_type: bigint
@@ -40,6 +46,12 @@ models:
         data_type: numeric
         description: "ratio of cancelled bookings over total bookings"
 
+      - name: is_cancellation_threshold_surpassed
+        data_type: boolean
+        description: "true if the cancellation ratio is higher than the set threshold"
+        tests:
+          - not_null
+
       - name: sum_cancelled_fee
         data_type: numeric
         description: "sum of fees charged for cancelled bookings

models/staging/edeposit/stg_edeposit__verifications.sql

@@ -27,7 +27,7 @@ with
     ),
     stg_edeposit__verifications as (
         select
-            {{ adapter.quote("documents") }} ->> 'id' as id,
+            {{ adapter.quote("documents") }} ->> 'id' as id_verification,
             {{ adapter.quote("documents") }} ->> 'BookingId' as id_booking,
             {{ adapter.quote("documents") }} ->> 'userId' as id_user,
             {{ adapter.quote("documents") }} ->> 'ListingId' as id_listing,