CRM Systems, Tools and Databases
The three system layers
Operational CRM runs daily transactions - sales force automation, service ticketing with a knowledge base, marketing automation. Analytical CRM stores and queries data for segmentation and scoring. Collaborative CRM links channels and partners.
Data structures
A database (OLTP) supports fast day-to-day transactions; a data warehouse (OLAP) holds historical, integrated data for analysis; a data mart is a subject-specific slice of the warehouse. Data moves by ETL (extract, transform, load). A primary key uniquely identifies a row; a foreign key points to a primary key in another table, creating a one-to-many relationship. Normalisation reduces duplication; a single customer view (golden record) merges duplicates into one reliable profile. A CDP unifies online and offline identifiers, which a classic CRM often cannot.
Worked example
Table customer(cust_id PK, name, mobile) and table complaint(ticket_id PK, cust_id FK, issue, status). To list every complaint for customer C-104, filter complaint rows where cust_id = C-104; the foreign key makes that join possible. If two rows carry the same mobile number under different cust_id values, you have duplicate profiles, not a shortcut.
Traps
- Confusing CRM with ERP - ERP handles finance, inventory and HR.
- Storing customer lists in spreadsheets without a unique ID, which breeds duplicates.
- Assuming cloud CRM is automatically secure and compliant; configuration and permissions decide that.
- Mistaking a data warehouse for a backup.