Shipping data model - cohort analysis development

models/marts/reporting/rep_shipping__cohort_analysis.sql

@@ -0,0 +1,212 @@
+-- SOURCE: https://www.holistics.io/blog/calculate-cohort-retention-analysis-with-sql/
+
+-- IMPORT
+WITH companies AS (
+    SELECT * FROM {{ ref('dim_shipping__companies') }}
+),
+
+origin_warehouse_shipping_packing_lists AS (
+    SELECT * FROM {{ ref('fct_shipping__origin_warehouse_shipping_packing_lists') }}
+),
+
+
+-- LOGIC
+company_register_month_year AS (
+
+    SELECT
+        companies.sub_company_id,
+        companies.sub_company_marking_id,
+        companies.sub_company_created_datetime,
+
+        IFF(
+            MAX(origin_warehouse_shipping_packing_lists.origin_warehouse_shipping_order_created_datetime) IS NULL,
+            1, 
+            0            
+        )   AS is_never_order_company,
+
+        DATE(DATE_TRUNC(
+            'MONTH', 
+            companies.sub_company_created_datetime
+        ))  AS register_month_year,
+
+        DATE(DATE_TRUNC(
+            'MONTH', 
+            MIN(origin_warehouse_shipping_packing_lists.origin_warehouse_shipping_order_created_datetime)
+        ))  AS first_order_month_year
+
+    FROM companies
+
+    LEFT JOIN origin_warehouse_shipping_packing_lists
+        ON (companies.sub_company_id = origin_warehouse_shipping_packing_lists.sub_company_id)
+
+    GROUP BY
+        companies.sub_company_id,
+        companies.sub_company_marking_id,
+        companies.sub_company_created_datetime,
+        register_month_year
+
+),
+
+company_order_count_from_registration_date AS (
+
+    SELECT
+        origin_warehouse_shipping_packing_lists.sub_company_id,
+
+        DATEDIFF(
+            MONTH, 
+            company_register_month_year.sub_company_created_datetime, 
+            origin_warehouse_shipping_packing_lists.origin_warehouse_shipping_order_created_datetime
+        )   AS order_month
+
+    FROM origin_warehouse_shipping_packing_lists
+
+    LEFT JOIN company_register_month_year
+        ON (origin_warehouse_shipping_packing_lists.sub_company_id = company_register_month_year.sub_company_id)
+    
+    GROUP BY
+        origin_warehouse_shipping_packing_lists.sub_company_id,
+        order_month
+
+),
+
+company_first_order_from_registration_date AS (
+
+    SELECT
+        origin_warehouse_shipping_packing_lists.sub_company_id,
+        DATE(DATE_TRUNC('MONTH', sub_company_created_datetime))     AS sub_company_created_datetime,
+        
+        MIN(DATEDIFF(
+            MONTH,
+            company_register_month_year.sub_company_created_datetime,
+            origin_warehouse_shipping_packing_lists.origin_warehouse_shipping_order_created_datetime
+        ))      AS first_order_month
+
+    FROM origin_warehouse_shipping_packing_lists
+
+    LEFT JOIN company_register_month_year
+        ON (origin_warehouse_shipping_packing_lists.sub_company_id = company_register_month_year.sub_company_id)
+
+    GROUP BY
+        origin_warehouse_shipping_packing_lists.sub_company_id,
+        sub_company_created_datetime
+
+),
+
+cohort_size_by_month_year AS (
+
+    SELECT
+        register_month_year,
+        SUM(is_never_order_company)     AS count_never_order_company,
+        COUNT(register_month_year)      AS count_total_company
+
+    FROM company_register_month_year
+
+    GROUP BY register_month_year
+
+    ORDER BY register_month_year
+
+),
+
+order_retention AS (
+    
+    SELECT
+        company_register_month_year.register_month_year,
+        company_order_count_from_registration_date.order_month,
+        COUNT(company_register_month_year.register_month_year)      AS count_retained_company
+
+        FROM company_order_count_from_registration_date
+
+        LEFT JOIN company_register_month_year
+            ON (company_order_count_from_registration_date.sub_company_id = company_register_month_year.sub_company_id)
+
+        GROUP BY
+            company_register_month_year.register_month_year,
+            company_order_count_from_registration_date.order_month
+
+),
+
+first_time_order_retention AS (
+
+    SELECT 
+        sub_company_created_datetime,
+        first_order_month,
+        COUNT(first_order_month)        AS count_first_order_company
+
+    FROM company_first_order_from_registration_date
+
+    GROUP BY 
+        sub_company_created_datetime,
+        first_order_month
+
+),
+
+cohort_analysis_table AS (
+
+    SELECT
+        order_retention.register_month_year,
+        order_retention.order_month,
+        order_retention.count_retained_company                    AS retained_company,
+
+        cohort_size_by_month_year.count_total_company             AS total_registered_company,
+        cohort_size_by_month_year.count_never_order_company,       
+
+        first_time_order_retention.count_first_order_company      AS count_first_order_company,
+
+        DIV0(
+            retained_company, 
+            total_registered_company                    
+        )       AS retention_rate_overall,
+
+        DIV0(
+            retained_company,
+            ( total_registered_company - cohort_size_by_month_year.count_never_order_company )
+        )       AS retention_rate_exclude_never_order_company,
+
+        DIV0(
+            count_first_order_company,
+            ( total_registered_company - count_never_order_company )
+        )       AS first_order_percentage,
+
+        '{{ modules.datetime.datetime.now(modules.pytz.timezone("Asia/Kuala_Lumpur")) }}'   AS _dbt_ran_datetime
+
+    FROM order_retention
+
+    LEFT JOIN cohort_size_by_month_year 
+        ON (order_retention.register_month_year = cohort_size_by_month_year.register_month_year)
+
+    LEFT JOIN first_time_order_retention
+        ON (order_retention.register_month_year = first_time_order_retention.sub_company_created_datetime)
+        AND (order_retention.order_month = first_time_order_retention.first_order_month)
+    
+    ORDER BY 
+        order_retention.register_month_year,
+        order_retention.order_month
+
+),
+
+
+-- FINAL
+final_rep_shipping__cohort_analysis AS (
+
+    SELECT
+        -- dimensions
+        register_month_year,
+        total_registered_company,
+        order_month,
+
+        -- measures
+        count_never_order_company,
+        count_first_order_company,
+        retained_company,
+        retention_rate_overall,
+        retention_rate_exclude_never_order_company,
+        first_order_percentage,
+
+        -- metadata
+        _dbt_ran_datetime
+    
+    FROM cohort_analysis_table
+
+)
+
+SELECT * FROM final_rep_shipping__cohort_analysis