Enterprises expend considerable effort to understand their customers and provide a superior experience and enhanced services to them, but still, they see an enormous difference between their understanding of customers’ requirements and their customers’ actual expectations.
This difference might be due to targeting the wrong customers, failing to understand their real requirements, lacking a convincing value proposition, or selecting the wrong channel for communication.
We propose to help handle these challenges with the help of a new framework that utilizes customer information and leverages suitable statistical models to help the decision support system to derive accurate customer intelligence.
This ‘Intelligence CRM’ framework will help customer-facing functions in effective decision support in order to improve the effectiveness of sales, marketing, and service functions across the enterprise, thereby resulting in improved customer satisfaction and retention.
Are Your Customer Interactions Being Guided by Intelligence?
Two important aspects of customer intelligence are “need” and “buying potential.” The former deals with what the customer might buy, and the latter with what the customer can potentially buy. Enterprises should understand their customers first, which means figuring out what products/services are important to them and the actual value they perceive in these products/services.
This customer data, if properly leveraged, can derive the necessary intelligence to help in understanding critical customer attributes — various aspects of customer behavior, customer buying potential, and customer lifetime value — thereby helping to increase the success rate of sales deals as well as order sizes.
This approach will be much more accurate and effective than the conventional practice because it is based on customer-related intelligence rather than just customer-related data.
As an example, a customer might have indicated the phone as the preferred communication channel, but an analysis of interaction and response history shows this customer responded effectively only to email communications. This intelligence can be leveraged to accurately identify the right communication channel with the customer, thus increasing the success rate of future interactions.
A CRM system should not only be operationally efficient but also provide the much-needed intelligence, which we call “Intelligence CRM” or “ICRM.”
The key building blocks of this system are problem identification, customer data capture, data collation, data mining, data analysis, and modeling. With increased global competition, varied product and service offerings from enterprises, and changing customer preferences, the need for ICRM will only increase.
Unearth the Hidden Information
Customer data collected by enterprises at various customer touchpoints — such as customer acquisition, order capture, billing, Web self-service, channel partners, or service request management — is maintained in several customer-facing/customer operations supporting systems. This data should be reconciled and refined by performing appropriate basic-to-complex data analysis in order to unearth relevant insights about the customer.
Imagine a CRM system that can provide meaningful insights about a customer’s propensity to churn or to buy a service/product offering — or reveal what that customer really values. In essence, we would need an “Intelligence CRM” system to provide this insight.
ICRM and Its Application
ICRM is a framework to derive intelligence and insights to address specific business challenges by applying configurable decision support models on data from multiple data sources spanning across the enterprise and beyond.
Traditionally, this approach has been leveraged on information available in CRM systems alone, thus limiting the quality of the intelligence/insight gained.
This might not be sufficient for the business to handle its challenges. With competition on the rise and customers getting more demanding, a 360-degree view is not sufficient to manage customers.
What businesses need today is a comprehensive “sphere of customer intelligence” gained from sources that go beyond CRM systems and the enterprise — that is, information from various sources such as billing systems, mediation systems, channel partners, analyst reports, social networking sites, etc.
ICRM leverages a three-layered structure to gain this intelligence: the data layer, the intelligence layer, and the usage layer, along with a modeling layer that cuts across all three.
The data layer is comprised of an enterprise-level online analytical processing (OLAP) system that gathers relevant information about the customer from various data sources and converts it into structured data.
The intelligence layer, which resides above the data layer, leverages business intelligence tools like real-time decisions and provides the first level of intelligence on customer attributes based on the criteria defined by the modeling layer.
The usage layer is the ICRM application layer, which is an empowered CRM application that pulls out information from the intelligence layer and provides the necessary insights to the business during customer interaction and decision support.
The modeling layer, which supports the other three layers, is the most critical layer. The modeling layer provides inputs to the data layer (enterprise OLAP) on the type of schema and attributes that are to be stored and takes the sample data for building models.
Technology and User Adoption
Any good business intelligence tool with a real-time decision system can help in building an effective decision-support system. However, there is no single system today that provides comprehensive insights into the business aspects by taking into consideration the individual insights from the respective sources discussed above.
This requires redefining the information architecture within an enterprise and going beyond the CRM system to leverage the customer information that resides in disparate systems across the enterprise in order to gain more meaningful insights with greater accuracy and relevance.
The ICRM architecture helps achieve this business need. Within the ICRM solution architecture, the ICRM system integrates with other legacy systems, external entities, decision support/intelligence tools, and the Enterprise OLAP.
Customer-related information should be passed on to the Enterprise OLAP through the CRM system, which has the provision to capture data from external entities that are the end-customer touchpoints, like channel partners, OEM partners, points of sale, and so on.
Any customer-related data stored in the legacy application databases is also passed to the enterprise OLAP. The enterprise OLAP structure is built so that data from several sources is collected in a structured format, as defined by modeling tools and intelligence tools.
Intelligence tools take logic and structure from modeling tools, as well as the data in a structured format from the Enterprise OLAP, and build and pass on the required insights — like churn prediction value or propensity to buy — to the ICRM system.
Customer insights are hidden in the form of data scattered across the enterprise in several source systems. A proper mechanism in the form of an “ICRM framework” devised to collect and process this information will derive valuable insights from it.
This is the “intelligence” that is required from a CRM system, which will help empower enterprises to interact with and serve their customers in a better-informed manner, building lasting and fruitful customer relationships.
Vamsi Krishna Paramjyothi is a consultant for enterprise solutions and Ramkumar Jonnalagadda is a principal consultant for enterprise solutions at Infosys Technologies.