specify the columns in the backfill procedures, makes sure the script…

…s are executable with a previous created feature table (#254)

sql/query/invoke_backfill_churn_propensity_label.sqlx

@@ -119,7 +119,13 @@ GROUP BY
 );
 
 -- Insert data into the target table, combining user information with churn and bounce status
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+  processed_timestamp,
+  feature_date,
+  user_pseudo_id,
+  churned,
+  bounced
+)
 SELECT DISTINCT
   -- Current timestamp as the processing timestamp
   CURRENT_TIMESTAMP() AS processed_timestamp,

sql/query/invoke_backfill_customer_lifetime_value_label.sqlx

@@ -109,7 +109,14 @@ CREATE OR REPLACE TEMP TABLE future_revenue_per_user AS (
 );
 
 -- Insert data into the target table
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+  processed_timestamp,
+  feature_date,
+  user_pseudo_id,
+  pltv_revenue_30_days,
+  pltv_revenue_90_days,
+  pltv_revenue_180_days
+)
   SELECT DISTINCT
     -- Current timestamp of the processing
     CURRENT_TIMESTAMP() AS processed_timestamp,

sql/query/invoke_backfill_purchase_propensity_label.sqlx

@@ -125,7 +125,26 @@ CREATE OR REPLACE TEMP TABLE future_purchases_per_user AS (
 );
 
 -- Inserts data into the target table
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+  processed_timestamp,
+  feature_date,
+  user_pseudo_id,
+  purchase_day_1,
+  purchase_day_2,
+  purchase_day_3,
+  purchase_day_4,
+  purchase_day_5,
+  purchase_day_6,
+  purchase_day_7,
+  purchase_day_8,
+  purchase_day_9,
+  purchase_day_10,
+  purchase_day_11,
+  purchase_day_12,
+  purchase_day_13,
+  purchase_day_14,
+  purchase_day_15_30
+)
 SELECT DISTINCT
   -- Selects the current timestamp and assigns it to the column processed_timestamp
   CURRENT_TIMESTAMP() AS processed_timestamp,

sql/query/invoke_backfill_user_dimensions.sqlx

@@ -122,7 +122,31 @@ CREATE OR REPLACE TEMP TABLE events_users as (
 ;
 
 -- Inserting aggregated user data into the target table.
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+    processed_timestamp,
+    feature_date,
+    user_pseudo_id,
+    user_id,
+    user_ltv_revenue,
+    device_category,
+    device_mobile_brand_name,
+    device_mobile_model_name,
+    device_os,
+    device_language,
+    device_web_browser,
+    geo_sub_continent,
+    geo_country,
+    geo_region,
+    geo_city,
+    geo_metro,
+    last_traffic_source_medium,
+    last_traffic_source_name,
+    last_traffic_source_source,
+    first_traffic_source_medium,
+    first_traffic_source_name,
+    first_traffic_source_source,
+    has_signed_in_with_user_id
+)
   SELECT DISTINCT
   -- Timestamp of the data processing
   CURRENT_TIMESTAMP() AS processed_timestamp,

sql/query/invoke_backfill_user_lifetime_dimensions.sqlx

@@ -137,7 +137,31 @@ CREATE OR REPLACE TEMP TABLE events_users as (
 -- This code block inserts data into the specified table, combining information from the "events_users" table
 -- and the "user_dimensions_event_session_scoped" table.
 -- It aggregates user-level features for each user and date.
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+  processed_timestamp,
+  feature_date,
+  user_pseudo_id,
+  user_id,
+  user_ltv_revenue,
+  device_category,
+  device_mobile_brand_name,
+  device_mobile_model_name,
+  device_os,
+  device_language,
+  device_web_browser,
+  geo_sub_continent,
+  geo_country,
+  geo_region,
+  geo_city,
+  geo_metro,
+  last_traffic_source_medium,
+  last_traffic_source_name,
+  last_traffic_source_source,
+  first_traffic_source_medium,
+  first_traffic_source_name,
+  first_traffic_source_source,
+  has_signed_in_with_user_id
+)
   SELECT DISTINCT
   -- The current timestamp.
   CURRENT_TIMESTAMP() AS processed_timestamp,

sql/query/invoke_backfill_user_lookback_metrics.sqlx

@@ -230,7 +230,25 @@ AND D.device_os IS NOT NULL
 -- This code is part of a larger process for building a machine learning model that predicts
 -- user behavior based on their past activity. The features generated by this code can be used
 -- as input to the model, helping it learn patterns and make predictions.
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+    processed_timestamp,
+    feature_date,
+    user_pseudo_id,
+    active_users_past_1_7_day,
+    active_users_past_8_14_day,
+    purchases_past_1_7_day,
+    purchases_past_8_14_day,
+    visits_past_1_7_day,
+    visits_past_8_14_day,
+    view_items_past_1_7_day,
+    view_items_past_8_14_day,
+    add_to_carts_past_1_7_day,
+    add_to_carts_past_8_14_day,
+    checkouts_past_1_7_day,
+    checkouts_past_8_14_day,
+    ltv_revenue_past_1_7_day,
+    ltv_revenue_past_7_15_day
+)
   SELECT DISTINCT
     -- Timestamp indicating when the data was processed
     CURRENT_TIMESTAMP() AS processed_timestamp,

sql/query/invoke_backfill_user_rolling_window_lifetime_metrics.sqlx

@@ -283,7 +283,50 @@ AND D.device_os IS NOT NULL
 -- This code is part of a larger process for building a machine learning model that predicts 
 -- user behavior based on their past activity. The features generated by this code can be used 
 -- as input to the model, helping it learn patterns and make predictions.
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+  processed_timestamp,
+  feature_date,
+  user_pseudo_id,
+  active_users_past_1_30_day,
+  active_users_past_30_60_day,
+  active_users_past_60_90_day,
+  active_users_past_90_120_day,
+  active_users_past_120_150_day,
+  active_users_past_150_180_day,
+  purchases_past_1_30_day,
+  purchases_past_30_60_day,
+  purchases_past_60_90_day,
+  purchases_past_90_120_day,
+  purchases_past_120_150_day,
+  purchases_past_150_180_day,
+  visits_past_1_30_day,
+  visits_past_30_60_day,
+  visits_past_60_90_day,
+  visits_past_90_120_day,
+  visits_past_120_150_day,
+  visits_past_150_180_day,
+  view_items_past_1_30_day,
+  view_items_past_30_60_day,
+  view_items_past_60_90_day,
+  view_items_past_90_120_day,
+  view_items_past_120_150_day,
+  view_items_past_150_180_day,
+  add_to_carts_past_1_30_day,
+  add_to_carts_past_30_60_day,
+  add_to_carts_past_60_90_day,
+  add_to_carts_past_90_120_day,
+  add_to_carts_past_120_150_day,
+  add_to_carts_past_150_180_day,
+  checkouts_past_1_30_day,
+  checkouts_past_30_60_day,
+  checkouts_past_60_90_day,
+  checkouts_past_90_120_day,
+  checkouts_past_120_150_day,
+  checkouts_past_150_180_day,
+  ltv_revenue_past_1_30_day,
+  ltv_revenue_past_30_90_day,
+  ltv_revenue_past_90_180_day
+)
 SELECT DISTINCT
   -- This selects the current timestamp and assigns it to the column processed_timestamp.
   CURRENT_TIMESTAMP() AS processed_timestamp,

sql/query/invoke_backfill_user_rolling_window_metrics.sqlx

@@ -272,7 +272,65 @@ CREATE OR REPLACE TEMP TABLE events_users as (
 -- table and several temporary tables containing rolling window features. The resulting data 
 -- represents user-level features for each user and date, capturing their past activity within 
 -- different time windows.
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+  processed_timestamp,
+  feature_date,
+  user_pseudo_id,
+  active_users_past_1_day,
+  active_users_past_2_day,
+  active_users_past_3_day,
+  active_users_past_4_day,
+  active_users_past_5_day,
+  active_users_past_6_day,
+  active_users_past_7_day,
+  active_users_past_8_14_day,
+  active_users_past_15_30_day,
+  purchases_past_1_day,
+  purchases_past_2_day,
+  purchases_past_3_day,
+  purchases_past_4_day,
+  purchases_past_5_day,
+  purchases_past_6_day,
+  purchases_past_7_day,
+  purchases_past_8_14_day,
+  purchases_past_15_30_day,
+  visits_past_1_day,
+  visits_past_2_day,
+  visits_past_3_day,
+  visits_past_4_day,
+  visits_past_5_day,
+  visits_past_6_day,
+  visits_past_7_day,
+  visits_past_8_14_day,
+  visits_past_15_30_day,
+  view_items_past_1_day,
+  view_items_past_2_day,
+  view_items_past_3_day,
+  view_items_past_4_day,
+  view_items_past_5_day,
+  view_items_past_6_day,
+  view_items_past_7_day,
+  view_items_past_8_14_day,
+  view_items_past_15_30_day,
+  add_to_carts_past_1_day,
+  add_to_carts_past_2_day,
+  add_to_carts_past_3_day,
+  add_to_carts_past_4_day,
+  add_to_carts_past_5_day,
+  add_to_carts_past_6_day,
+  add_to_carts_past_7_day,
+  add_to_carts_past_8_14_day,
+  add_to_carts_past_15_30_day,
+  checkouts_past_1_day,
+  checkouts_past_2_day,
+  checkouts_past_3_day,
+  checkouts_past_4_day,
+  checkouts_past_5_day,
+  checkouts_past_6_day,
+  checkouts_past_7_day,
+  checkouts_past_8_14_day,
+  checkouts_past_15_30_day
+)
   SELECT DISTINCT
     -- This selects the current timestamp and assigns it to the column processed_timestamp.
     CURRENT_TIMESTAMP() AS processed_timestamp,

sql/query/invoke_backfill_user_scoped_lifetime_metrics.sqlx

@@ -163,7 +163,35 @@ CREATE OR REPLACE TEMP TABLE first_purchasers as (
 );
 
 -- This SQL code calculates various user engagement and revenue metrics at a daily level and inserts the results into a target table. It leverages several temporary tables created earlier in the script to aggregate data efficiently.
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+  processed_timestamp,
+  feature_date,
+  lifetime_purchasers_users,
+  lifetime_average_daily_purchasers,
+  lifetime_active_users,
+  lifetime_DAU,
+  lifetime_MAU,
+  lifetime_WAU,
+  lifetime_dau_per_mau,
+  lifetime_dau_per_wau,
+  lifetime_wau_per_mau,
+  lifetime_users_engagement_duration_seconds,
+  lifetime_average_engagement_time,
+  lifetime_average_engagement_time_per_session,
+  lifetime_average_sessions_per_user,
+  lifetime_ARPPU,
+  lifetime_ARPU,
+  lifetime_average_daily_revenue,
+  lifetime_max_daily_revenue,
+  lifetime_min_daily_revenue,
+  lifetime_new_users,
+  lifetime_returning_users,
+  lifetime_first_time_purchasers,
+  lifetime_first_time_purchaser_conversion,
+  lifetime_first_time_purchasers_per_new_user,
+  lifetime_avg_user_conversion_rate,
+  lifetime_avg_session_conversion_rate
+)
 SELECT
  -- Records the current timestamp when the query is executed.
  CURRENT_TIMESTAMP() AS processed_timestamp,

sql/query/invoke_backfill_user_scoped_metrics.sqlx

@@ -183,7 +183,35 @@ CREATE OR REPLACE TEMP TABLE new_users_ as (
 );
 
 -- Insert data into the target table after calculating various user engagement and revenue metrics.
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+    processed_timestamp,
+    feature_date,
+    purchasers_users,
+    average_daily_purchasers,
+    active_users,
+    DAU,
+    MAU,
+    WAU,
+    dau_per_mau,
+    dau_per_wau,
+    wau_per_mau,
+    users_engagement_duration_seconds,
+    average_engagement_time,
+    average_engagement_time_per_session,
+    average_sessions_per_user,
+    ARPPU,
+    ARPU,
+    average_daily_revenue,
+    max_daily_revenue,
+    min_daily_revenue,
+    new_users,
+    returning_users,
+    first_time_purchasers,
+    first_time_purchaser_conversion,
+    first_time_purchasers_per_new_user,
+    avg_user_conversion_rate,
+    avg_session_conversion_rate
+)
   SELECT DISTINCT
     -- Record the current timestamp when the query is executed.
     CURRENT_TIMESTAMP() AS processed_timestamp,

sql/query/invoke_backfill_user_scoped_segmentation_metrics.sqlx

@@ -136,7 +136,35 @@ GROUP BY feature_date
 );
 
 -- This SQL code calculates various user engagement and revenue metrics at a daily level and inserts the results into a target table. It leverages several temporary tables created earlier in the script to aggregate data efficiently.
-INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}`
+INSERT INTO `{{project_id}}.{{dataset}}.{{insert_table}}` (
+  processed_timestamp,
+  feature_date,
+  purchasers_users,
+  average_daily_purchasers,
+  active_users,
+  DAU,
+  MAU,
+  WAU,
+  dau_per_mau,
+  dau_per_wau,
+  wau_per_mau,
+  users_engagement_duration_seconds,
+  average_engagement_time,
+  average_engagement_time_per_session,
+  average_sessions_per_user,
+  ARPPU,
+  ARPU,
+  average_daily_revenue,
+  max_daily_revenue,
+  min_daily_revenue,
+  new_users,
+  returning_users,
+  first_time_purchasers,
+  first_time_purchaser_conversion,
+  first_time_purchasers_per_new_user,
+  avg_user_conversion_rate,
+  avg_session_conversion_rate
+)
   SELECT
     -- Records the current timestamp when the query is executed.
     CURRENT_TIMESTAMP() AS processed_timestamp,