Deng 2579 gclid conversions v2 creation (#4972)

* initial commit of code for gclid_conversions_v2

* change owners to Marlene and Katie

* add gclid_conversions_v2 to Access Denied bqetl_project.yaml

bqetl_project.yaml

@@ -190,6 +190,7 @@ dry_run:
   - sql/moz-fx-data-shared-prod/stub_attribution_service_derived/dl_token_ga_attribution_lookup_v1/query.sql
   - sql/moz-fx-data-shared-prod/stub_attribution_service/dl_token_ga_attribution_lookup/view.sql
   - sql/moz-fx-data-shared-prod/mozilla_org_derived/gclid_conversions_v1/query.sql
+  - sql/moz-fx-data-shared-prod/mozilla_org_derived/gclid_conversions_v2/query.sql
   - sql/moz-fx-data-shared-prod/mozilla_org/gclid_conversions/view.sql
   # Materialized views
   - sql/moz-fx-data-shared-prod/telemetry_derived/experiment_search_events_live_v1/init.sql

sql/moz-fx-data-shared-prod/mozilla_org_derived/gclid_conversions_v2/checks.sql

@@ -0,0 +1,2 @@
+#fail
+{{ is_unique(["gclid", "activity_date"]) }}

sql/moz-fx-data-shared-prod/mozilla_org_derived/gclid_conversions_v2/metadata.yaml

@@ -0,0 +1,28 @@
+friendly_name: Gclid Conversions
+description: |-
+  All conversions for GCLIDs.
+  Partitioned by activity_date; a single GCLID
+  can have multiple entries in this table,
+  if they perform different conversion actions
+  on different days.
+owners:
+- mhirose@mozilla.com
+- kwindau@mozilla.com
+labels:
+  incremental: true
+  owner1: mhirose@mozilla.com
+scheduling:
+  dag_name: bqetl_mozilla_org_derived
+  date_partition_parameter: activity_date
+  parameters: ["conversion_window:INT64:30"]
+bigquery:
+  time_partitioning:
+    type: day
+    field: 'activity_date'
+    require_partition_filter: false
+    expiration_days: null
+references: {}
+deprecated: false
+workgroup_access:
+- role: roles/bigquery.dataViewer
+  members: [workgroup:dataops-managed/external-census]

sql/moz-fx-data-shared-prod/mozilla_org_derived/gclid_conversions_v2/query.sql

@@ -0,0 +1,71 @@
+WITH gclids_to_ga_ids AS (
+  SELECT
+    gclid,
+    ga_client_id,
+    stub_session_id,
+  FROM
+    mozilla_org_derived.ga_sessions_v2
+  CROSS JOIN
+    UNNEST(all_reported_stub_session_ids) AS stub_session_id
+  WHERE
+    session_date >= DATE_SUB(@activity_date, INTERVAL @conversion_window DAY)
+    AND gclid IS NOT NULL
+),
+ga_ids_to_dl_token AS (
+  SELECT
+    ga_client_id,
+    stub_session_id,
+    dl_token,
+  FROM
+    stub_attribution_service_derived.dl_token_ga_attribution_lookup_v1
+  WHERE
+    ga_client_id IS NOT NULL
+    AND stub_session_id IS NOT NULL
+),
+dl_token_to_telemetry_id AS (
+  SELECT
+    client_id AS telemetry_client_id,
+    first_seen_date,
+    attribution_dltoken AS dl_token,
+  FROM
+    telemetry_derived.clients_first_seen_v2
+),
+telemetry_id_to_activity AS (
+  SELECT
+    client_id AS telemetry_client_id,
+    submission_date AS activity_date,
+    search_count_all > 0 AS did_search,
+    ad_clicks_count_all > 0 AS did_click_ad,
+    TRUE AS was_active,
+  FROM
+    telemetry_derived.clients_daily_v6
+  WHERE
+    submission_date = @activity_date
+)
+SELECT
+  activity_date,
+  gclid,
+  COALESCE(
+    LOGICAL_OR(was_active AND activity_date = first_seen_date),
+    FALSE
+  ) AS did_firefox_first_run,
+  COALESCE(LOGICAL_OR(did_search), FALSE) AS did_search,
+  COALESCE(LOGICAL_OR(did_click_ad), FALSE) AS did_click_ad,
+  COALESCE(
+    LOGICAL_OR(was_active AND activity_date > first_seen_date),
+    FALSE
+  ) AS did_returned_second_day,
+FROM
+  gclids_to_ga_ids
+INNER JOIN
+  ga_ids_to_dl_token
+  USING (ga_client_id, stub_session_id)
+INNER JOIN
+  dl_token_to_telemetry_id
+  USING (dl_token)
+INNER JOIN
+  telemetry_id_to_activity
+  USING (telemetry_client_id)
+GROUP BY
+  activity_date,
+  gclid

sql/moz-fx-data-shared-prod/mozilla_org_derived/gclid_conversions_v2/schema.yaml

@@ -0,0 +1,25 @@
+fields:
+  - name: activity_date
+    mode: NULLABLE
+    type: DATE
+    description: "The date that the conversion action took place."
+  - name: gclid
+    mode: NULLABLE
+    type: STRING
+    description: "A Google Click ID, which uniquely represent an ad impression for Google ads."
+  - name: did_firefox_first_run
+    mode: NULLABLE
+    type: BOOLEAN
+    description: "Whether the GA session associated with this GCLID resulted in a Firefox install which ran the first time on this day."
+  - name: did_search
+    mode: NULLABLE
+    type: BOOLEAN
+    description: "Whether the GA session associated with this GCLID resulted in a Firefox install with a search on this day."
+  - name: did_click_ad
+    mode: NULLABLE
+    type: BOOLEAN
+    description: "Whether the GA session associated with this GCLID resulted in a Firefox install with an ad click on this day."
+  - name: did_returned_second_day
+    mode: NULLABLE
+    type: BOOLEAN
+    description: "Whether the GA session associated with this GCLID resulted in a Firefox install that returned on this day."