Lead score propensity use case for non-ecommerce

infrastructure/terraform/README.md

@@ -106,6 +106,7 @@ Also, this method allows you to extend this solution and develop it to satisfy y
     Terraform stores state about managed infrastructure to map real-world resources to the configuration, keep track of metadata, and improve performance. Terraform stores this state in a local file by default, but you can also use a Terraform remote backend to store state remotely. [Remote state](https://developer.hashicorp.com/terraform/cdktf/concepts/remote-backends) makes it easier for teams to work together because all members have access to the latest state data in the remote store.
 
     ```bash
+    SOURCE_ROOT="${HOME}/${REPO}"
     cd ${SOURCE_ROOT}
     scripts/generate-tf-backend.sh
     ```

infrastructure/terraform/modules/activation/export-procedures.tf

@@ -119,10 +119,33 @@ resource "google_bigquery_routine" "export_churn_propensity_procedure" {
   routine_type    = "PROCEDURE"
   language        = "SQL"
   definition_body = data.template_file.churn_propensity_csv_export_query.rendered
-  description     = "Export purchase propensity predictions as CSV for GA4 User Data Import"
+  description     = "Export churn propensity predictions as CSV for GA4 User Data Import"
   arguments {
     name      = "prediction_table_name"
     mode      = "IN"
     data_type = jsonencode({ "typeKind" : "STRING" })
   }
 }
+
+data "template_file" "lead_score_propensity_csv_export_query" {
+  template = file("${local.source_root_dir}/templates/activation_user_import/lead_score_propensity_csv_export.sqlx")
+  vars = {
+    ga4_stream_id = var.ga4_stream_id
+    export_bucket = module.pipeline_bucket.name
+  }
+}
+
+resource "google_bigquery_routine" "export_lead_score_propensity_procedure" {
+  project         = null_resource.check_bigquery_api.id != "" ? module.project_services.project_id : var.project_id
+  dataset_id      = module.bigquery.bigquery_dataset.dataset_id
+  routine_id      = "export_lead_score_propensity_predictions"
+  routine_type    = "PROCEDURE"
+  language        = "SQL"
+  definition_body = data.template_file.lead_score_propensity_csv_export_query.rendered
+  description     = "Export lead score propensity predictions as CSV for GA4 User Data Import"
+  arguments {
+    name      = "prediction_table_name"
+    mode      = "IN"
+    data_type = jsonencode({ "typeKind" : "STRING" })
+  }
+}

infrastructure/terraform/modules/activation/main.tf

@@ -24,6 +24,8 @@ locals {
   cltv_query_template_file                       = "cltv_query_template.sqlx"
   purchase_propensity_query_template_file        = "purchase_propensity_query_template.sqlx"
   purchase_propensity_vbb_query_template_file    = "purchase_propensity_vbb_query_template.sqlx"
+  lead_score_propensity_query_template_file      = "lead_score_propensity_query_template.sqlx"
+  lead_score_propensity_vbb_query_template_file  = "lead_score_propensity_vbb_query_template.sqlx"
   churn_propensity_query_template_file           = "churn_propensity_query_template.sqlx"
   activation_container_image_id                  = "activation-pipeline"
   docker_repo_prefix                             = "${var.location}-docker.pkg.dev/${var.project_id}"
@@ -750,7 +752,7 @@ data "template_file" "churn_propensity_query_template_file" {
   }
 }
 
-# This resource creates a bucket object using as content the purchase_propensity_query_template_file file.
+# This resource creates a bucket object using as content the churn_propensity_query_template_file file.
 resource "google_storage_bucket_object" "churn_propensity_query_template_file" {
   name    = "${local.configuration_folder}/${local.churn_propensity_query_template_file}"
   content = data.template_file.churn_propensity_query_template_file.rendered
@@ -791,6 +793,40 @@ resource "google_storage_bucket_object" "purchase_propensity_vbb_query_template_
   bucket  = module.pipeline_bucket.name
 }
 
+data "template_file" "lead_score_propensity_query_template_file" {
+  template = file("${local.template_dir}/activation_query/${local.lead_score_propensity_query_template_file}")
+
+  vars = {
+    mds_project_id     = var.mds_project_id
+    mds_dataset_suffix = var.mds_dataset_suffix
+  }
+}
+
+# This resource creates a bucket object using as content the lead_score_propensity_query_template_file file.
+resource "google_storage_bucket_object" "lead_score_propensity_query_template_file" {
+  name    = "${local.configuration_folder}/${local.lead_score_propensity_query_template_file}"
+  content = data.template_file.lead_score_propensity_query_template_file.rendered
+  bucket  = module.pipeline_bucket.name
+}
+
+# This resource creates a bucket object using as content the lead_score_propensity_vbb_query_template_file file.
+data "template_file" "lead_score_propensity_vbb_query_template_file" {
+  template = file("${local.template_dir}/activation_query/${local.lead_score_propensity_vbb_query_template_file}")
+
+  vars = {
+    mds_project_id        = var.mds_project_id
+    mds_dataset_suffix    = var.mds_dataset_suffix
+    activation_project_id = var.project_id
+    dataset               = module.bigquery.bigquery_dataset.dataset_id
+  }
+}
+
+resource "google_storage_bucket_object" "lead_score_propensity_vbb_query_template_file" {
+  name    = "${local.configuration_folder}/${local.lead_score_propensity_vbb_query_template_file}"
+  content = data.template_file.lead_score_propensity_vbb_query_template_file.rendered
+  bucket  = module.pipeline_bucket.name
+}
+
 # This data resources creates a data resource that renders a template file and stores the rendered content in a variable.
 data "template_file" "activation_type_configuration" {
   template = file("${local.template_dir}/activation_type_configuration_template.tpl")
@@ -802,6 +838,8 @@ data "template_file" "activation_type_configuration" {
     purchase_propensity_query_template_gcs_path        = "gs://${module.pipeline_bucket.name}/${google_storage_bucket_object.purchase_propensity_query_template_file.output_name}"
     purchase_propensity_vbb_query_template_gcs_path    = "gs://${module.pipeline_bucket.name}/${google_storage_bucket_object.purchase_propensity_vbb_query_template_file.output_name}"
     churn_propensity_query_template_gcs_path           = "gs://${module.pipeline_bucket.name}/${google_storage_bucket_object.churn_propensity_query_template_file.output_name}"
+    lead_score_propensity_query_template_gcs_path        = "gs://${module.pipeline_bucket.name}/${google_storage_bucket_object.lead_score_propensity_query_template_file.output_name}"
+    lead_score_propensity_vbb_query_template_gcs_path    = "gs://${module.pipeline_bucket.name}/${google_storage_bucket_object.lead_score_propensity_vbb_query_template_file.output_name}"
   }
 }
 

infrastructure/terraform/modules/feature-store/bigquery-datasets.tf

@@ -90,6 +90,32 @@ resource "google_bigquery_dataset" "churn_propensity" {
   }
 }
 
+# This resource creates a BigQuery dataset called `lead_score_propensity`.
+resource "google_bigquery_dataset" "lead_score_propensity" {
+  dataset_id    = local.config_bigquery.dataset.lead_score_propensity.name
+  friendly_name = local.config_bigquery.dataset.lead_score_propensity.friendly_name
+  project       = null_resource.check_bigquery_api.id != "" ? local.lead_score_propensity_project_id : local.feature_store_project_id
+  description   = local.config_bigquery.dataset.lead_score_propensity.description
+  location      = local.config_bigquery.dataset.lead_score_propensity.location
+  # The max_time_travel_hours attribute specifies the maximum number of hours that data in the dataset can be accessed using time travel queries. 
+  # In this case, the maximum time travel hours is set to the value of the local file config.yaml section bigquery.dataset.feature_store.max_time_travel_hours configuration.
+  max_time_travel_hours = local.config_bigquery.dataset.lead_score_propensity.max_time_travel_hours
+  # The delete_contents_on_destroy attribute specifies whether the contents of the dataset should be deleted when the dataset is destroyed. 
+  # In this case, the delete_contents_on_destroy attribute is set to false, which means that the contents of the dataset will not be deleted when the dataset is destroyed.
+  delete_contents_on_destroy = false
+
+  labels = {
+    version = "prod"
+  }
+
+  # The lifecycle block allows you to configure the lifecycle of the dataset. 
+  # In this case, the ignore_changes attribute is set to all, which means that 
+  # Terraform will ignore any changes to the dataset and will not attempt to update the dataset.
+  lifecycle {
+    ignore_changes = all
+  }
+}
+
 # This resource creates a BigQuery dataset called `customer_lifetime_value`.
 resource "google_bigquery_dataset" "customer_lifetime_value" {
   dataset_id    = local.config_bigquery.dataset.customer_lifetime_value.name
@@ -300,7 +326,8 @@ module "gemini_insights" {
   location     = local.config_bigquery.dataset.gemini_insights.location
   # The delete_contents_on_destroy attribute specifies whether the contents of the dataset should be deleted when the dataset is destroyed. 
   # In this case, the delete_contents_on_destroy attribute is set to false, which means that the contents of the dataset will not be deleted when the dataset is destroyed.
-  delete_contents_on_destroy = true
+  delete_contents_on_destroy = false
+  deletion_protection = true
 
   dataset_labels = {
     version    = "prod",
@@ -314,7 +341,7 @@ module "gemini_insights" {
       # The max_time_travel_hours attribute specifies the maximum number of hours that data in the dataset can be accessed using time travel queries.
       # In this case, the maximum time travel hours is set to the value of the local file config.yaml section bigquery.dataset.gemini_insights.max_time_travel_hours configuration.
       max_time_travel_hours = local.config_bigquery.dataset.gemini_insights.max_time_travel_hours
-      deletion_protection   = false
+      deletion_protection   = true
       time_partitioning     = null,
       range_partitioning    = null,
       expiration_time       = null,