By combining text and data mining technologies, CLEMCRM can be applied in a wide range of areas from customer retention to marketing cost reduction to personalised customer experience.

Customer segmentation
Based on the customer’s past transactions, their personal and behavioural data specific customer segments can be created, which includes similar customers. Using the segmentation performed by CLEMCRM the main segments and patterns of the customers can be identified. Applying the customer profiles helps organizations to support the development of strategy and marketing campaigns: every customer can be reached with different offers on different ways with various promotions, discounts to increase their satisfaction.
Customer segments may serve as the basis of further analyses, for example customer value, customer retention, increase customer number, cross-selling and fraud detection.

Churn Forecasting
Applying CLEMCRM organizations can estimate the probability of customer's churn for the next period based on previous customer, transaction and behavioural data. CLEMCRM is a predictive analytics solution, which reveals the churn behaviour and provides deeper insight into the customer's reasons. After identifying risky customers, using the appropriate steps organizations can prevent their churn and strengthen their loyalty. Customer segmentation can help to identify the right next steps.

Customer Value
Based on previous customer, transaction and behavioural data, applying CLEMCRM's predictive methods organizations can classify customers according to their future profitability and probability to churn. Using CLEMCRM the expected profit from every customer can be predicted so personalized customer experience is provided to every customer. The segmentation supports the identification of specific customer needs, so client's with high customer value can easily be targeted with marketing actions to strengthen the long-term loyalty.

Algorithms
CLEMCRM uses modelling algorithms from the wide model palette of IBM SPSS Modeler: from the popular regression models to the decision trees and neural networks, there are many segmentation and classification models. CLEMCRM solution includes a specific module based on a co-clustering algorithm to develop more accurate customer segmentation and cross-selling models to improves the efficiency of the marketing campaigns.

Evaluation
The completed models are validated with the usual methodology of data mining tools: hit matrix, gain, lift, ROI.

Automation
It is important to integrate and automate the solution into everyday business processes, from data transformation to automated reporting and customer classification.