Merge pull request #19 from Ritik574-coder/dbt_branch

feat(intermediate): build customer profile, location, contact, and segmentation models

Author: Ritik574-coder

models/intermediate/Customers/int_customer_location.sql

@@ -1,14 +1,17 @@
 -- ============================================================================
 -- Model: int_customer_location
-
+--
 -- Purpose:
---   Standardizes customer location attributes by correcting common data quality
---   issues such as missing values, typographical errors, inconsistent casing,
---   invalid state abbreviations, invalid ZIP codes, and non-standard country
---   names. This model prepares clean and consistent geographic information for
---   downstream analytical models.
-
--- Maintainer: Ritik   
+--   Creates a trusted customer location dataset by standardizing geographic
+--   attributes and resolving common data quality issues across customer
+--   addresses. This model corrects city name spelling errors, normalizes
+--   address formatting, validates state and ZIP code information, and
+--   standardizes country values into business-defined categories. The
+--   resulting dataset provides reliable geographic attributes for customer
+--   segmentation, regional reporting, sales analysis, and downstream
+--   dimensional models.
+--
+-- Maintainer: Ritik
 -- ============================================================================
 
 -- dbt.TitleCase is a  custom function taht help me to convert string into title case 

models/intermediate/Customers/int_customer_profile.sql

@@ -2,11 +2,13 @@
 -- Model: int_customer_profile
 --
 -- Purpose:
---   Standardizes customer profile attributes by cleaning and transforming
---   personal information such as titles, names, gender, and date of birth.
---   This model normalizes inconsistent values, extracts first and last names,
---   standardizes gender categories, and converts multiple date formats into a
---   consistent DATE data type for reliable downstream analytical models.
+--   Creates a standardized customer profile dataset by applying business
+--   rules to personal customer attributes. This model derives first and last
+--   names from the full name, normalizes gender values into consistent
+--   categories, and converts multiple source date formats into a reliable
+--   DATE data type. The resulting dataset serves as a trusted foundation for
+--   customer demographics, segmentation, personalization, and downstream
+--   dimensional models.
 --
 -- Maintainer: Ritik
 -- ============================================================================

models/intermediate/Customers/int_customer_segmentation.sql

@@ -1,9 +1,57 @@
+-- ============================================================================
+-- Model: int_customer_segmentation
+--
+-- Purpose:
+--   Creates a standardized customer segmentation dataset by applying business
+--   rules to customer profile and account attributes. This model normalizes
+--   customer segments, imputes missing annual income using segment-level
+--   median values, standardizes company names and account status values, and
+--   converts inconsistent account creation dates into a reliable DATE format.
+--   The resulting dataset serves as a trusted foundation for customer
+--   segmentation, loyalty analysis, marketing analytics, and dimensional
+--   modeling.
+--
+-- Maintainer: Ritik
+-- ============================================================================
 SELECT
     customer_id,
-    customer_segment,
-    loyalty_points,
-    annual_income_usd,
-    company,
-    is_active,
-    account_created_date
+    CASE 
+        WHEN customer_segment IS NULL OR customer_segment = '' THEN 'Unknown' 
+        ELSE customer_segment
+    END as customer_segment, -- distinct value is Bronze Gold Platinum Silver
+
+    TRY_CAST(loyalty_points AS INT) as loyalty_points,
+
+    COALESCE(
+        annual_income_usd,
+        PERCENTILE_CONT(0.5)
+        WITHIN GROUP (ORDER BY annual_income_usd)
+        OVER (PARTITION BY customer_segment)
+    ) as annual_income_usd,
+
+    CASE 
+        WHEN company IS NULL OR company = '' THEN 'Unknown'
+        WHEN TRIM(REPLACE(REPLACE(company, CHAR(13), ''), CHAR(10), '')) = '' THEN 'Unknown'
+        ELSE TRIM(REPLACE(REPLACE(company, CHAR(13), ''), CHAR(10), ''))
+    END as company,
+
+    CASE 
+         WHEN TRIM(LOWER(is_active)) IN ('1', 'active', 'true', 'y', 'yes')   THEN 'True'
+         WHEN TRIM(LOWER(is_active)) IN ('0', 'inactive', 'false', 'n', 'no') THEN 'False'
+         ELSE 'Unknown'
+    END AS is_active,
+
+    CASE
+        WHEN TRIM(account_created_date) LIKE '[A-Z][a-z][a-z] __, ____'       THEN CONVERT(DATE,account_created_date)
+        WHEN TRIM(account_created_date) LIKE '[A-Z][a-z][a-z][a-z]% __, ____' THEN CONVERT(DATE,account_created_date)
+        WHEN TRIM(account_created_date) LIKE '____/__/__'                     THEN CONVERT(DATE,account_created_date)
+        WHEN TRIM(account_created_date) LIKE '____-__-__'                     THEN CONVERT(DATE,account_created_date)
+
+        WHEN TRIM(account_created_date) LIKE '__-__-____' AND TRY_CONVERT(INT, LEFT(account_created_date,2)) > 12          THEN TRY_CONVERT(DATE, account_created_date, 105)
+        WHEN TRIM(account_created_date) LIKE '__-__-____' AND TRY_CONVERT(INT, SUBSTRING(account_created_date,4,2)) > 12   THEN TRY_CONVERT(DATE, account_created_date, 110)
+        WHEN TRIM(account_created_date) LIKE '__/__/____' AND TRY_CONVERT(INT, LEFT(account_created_date,2)) > 12          THEN TRY_CONVERT(DATE, account_created_date, 103)
+        WHEN TRIM(account_created_date) LIKE '__/__/____' AND TRY_CONVERT(INT, SUBSTRING(account_created_date, 4, 2)) > 12 THEN TRY_CONVERT(DATE, account_created_date, 101)
+        ELSE TRY_CONVERT(DATE, account_created_date,101)
+    END as account_created_date
+
 FROM {{ ref('stg_customers') }} ;

models/intermediate/Customers/int_customers_contact.sql

@@ -2,11 +2,13 @@
 -- Model: int_customers_contact
 --
 -- Purpose:
---   Standardizes customer contact information by cleaning and validating
---   email addresses, phone numbers, and preferred communication channels.
---   This model corrects common email domain typographical errors, formats
---   phone numbers into a consistent structure, and normalizes customer
---   communication preferences for reliable downstream analytical models.
+--   Creates a trusted customer contact dataset by validating, correcting, and
+--   standardizing customer communication attributes. This model repairs common
+--   email formatting issues and domain typographical errors, transforms phone
+--   numbers into a consistent business format, and normalizes preferred
+--   communication channels into standardized business categories. The resulting
+--   dataset provides reliable contact information for customer communication,
+--   marketing campaigns, CRM operations, and downstream analytical models.
 --
 -- Maintainer: Ritik
 -- ============================================================================