Analytical CRM is the part of Customer Relationship Management that aims at storing, analyzing and applying the knowledge about customers and about ways to approach customers, typically using databases, statistical tools, data mining, machine learning, Business Intelligence and reporting methodologies. Analytical CRM analyses data (gathered as part of operational CRM, or from other sources) in an attempt to identify means to enhance a company’s relationship with its clients.
Customer knowledge consists out of:
- Basic personal data such as: customer name, company name, business unit, business department, address, email, phone, fax, gender, nationality, etc
- More sophisticated client knowledge such as:
- Client value (annual revenue, profitability)
- Transactions (product description, revenue, profit, payment method, payment behavior)
- Internet communication (IP-address, entry page, click stream, visit length)
- Telephone communication (call center report data, sales calls)
- Other communications (mailings, response)
- Customer satisfaction (with product, service, company)
This client information can be captured from the processes (sales, services, finance, marketing) and channels (Multi Channel Marketing) of the organization. Certain data can also be acquired from external sources, such as market research data or address databases. It is often advisable to store all client data centrally for the organization to avoid ‘multiple versions of the truth’. Client data should be actual, complete, correct, unique (each client should be in the database only 1 time) and accessible for those who need it when they need it. This is true a fortiori for companies with a strategic Customer Relationship Management philosophy.
Usage of Analytical CRM
The results of an analysis can be used to design targeted marketing campaigns, for example:
- Acquisition: Cross-selling, up-selling
- Retention: Retaining existing customers (antonym: customer attrition)
- Information: Providing timely and regular information to customers
Other examples of the applications of analyses include:
- Contact optimization
- Evaluating and improving customer satisfaction
- Optimizing sales coverage
- Fraud detection
- Financial forecasts
- Price optimization
- Product development
- Program evaluation
- Risk assessment and management
- Strategic Marketing
- Operational marketing
Steps in Analytical CRM
After the client data is collected and stored, the actual analysis can take place. The analysis process is roughly made of the following steps:
- Problem formulation. What do we want to know. Is answering the question relevant and possible (technically, financially and organizationally). Typical a CRM analysis question is about:
- Segmentation of clients
- Acquisition analysis (what is the quality of various lists or databases)
- Relation analysis (expected retention, opportunities for cross-selling, deep-selling, up-selling)
- Channel or approach analysis (which channel or approach gives the best results)
- Preparation (random sample survey, relevant variables, cases, spread in scores, prepare definitive dataset)
- Definitive analysis, using:
- Statistical techniques (Regression Analysis, Dynamic Regression, Exploratory Factor Analysis, Exponential Smoothing, ARIMA)
- Datamining, typically aimed at discovering non-obvious or non-linear patterns in the data.
- Machine learning (Artificial Intelligence) techniques, such as: neural networks, genetic algorithms, association rules and case-based reasoning.
- Visualizing the results in such a way that it is understandable for the users.
Strengths of Analytical CRM
- Can help to find and explore useful knowledge in large customer databases.
- Classify customers, predict customer behavior, select market approach or channel.
Limitations of Analytical CRM
- Certain Analytical CRM techniques can be complex to understand.
- Still in early stage of usage. .