Customer Data Intergration Software

 Customer Data Integration: Creating a Single Version of the Truth

Effective customer management happens with online, accurate, integrated and up-to-date customer information. In most organizations, customer information is distributed and duplicated across various applications, and it is difficult to get a single version of the truth. To enhance customer management, organizations are increasingly investing time and money into customer data management, with the customer data integration (CDI) system. CDI involves the integrating and unifying of customer information from disperse and heterogeneous business applications. This integrated customer repository becomes the central repository of customer information being used by various applications, and produces the true view of the customer. 

The customer is the heart and soul of an organization. This is the central entity around which business of an organization revolves. The customer is the golden nugget on which the survival and prosperity of an organization depends. The entire universe may be considered as the customer base of an organization in a true sense. The customer can be viewed in various perspectives, three of which are shown in Figure 1.



Figure 1: Customer Classification
From the business interaction angle, customers can be:

• Account holders: The existing customers possessing active, inactive, dormant or closed accounts.
• Organizations: The internal organization can be analyzed at various levels of granularity and functions, e.g., the employees themselves can be account holders.
• Partners: These are channels as well as business supporters. Partner behavior can be studied to understand their needs through the business events. This can be effectively utilized to earn loyalty as well as to grow business.
• Competitors: In order to strategically frame the business, competitor activities need to be monitored.

Based on status, the customers can be classified as:

• Active customer: One whose account is active and running.
• Inactive customer: Customer having valid account but has not used in a period of time.
• Dormant customer: Individual or corporation with whom organization did business in the past.
• Prospective customer: An individual or corporate that can be targeted as a potential customer.

Based on the organization, the customer can be grouped as:

• Individual: A specific person of interest to the organization, i.e., employee, agent or dealer.
• Corporate: A group of individuals who have banded together for a commercial purpose.

CDI Overview

Organizations have to build strong customer relationships to stay competitive and grow in today's market. Effective customer management mandates easy and quick access to up-to-date and accurate customer information. To enhance customer management, organizations are investing time and money toward building an integrated central customer repository, which can provide the online, accurate, integrated and up-to-date customer information.
Customer data integration (CDI) is an approach toward integration and unification of the customer information from disperse and heterogeneous business applications. CDI processes consolidate customer information from all available sources, such as operational systems, call centers, customer relationship management (CRM) and data warehousing (DW) applications, and ensures the access of the current and complete view of customer information to the relevant departments/ business groups.



Figure 2: CDI Context Diagram
A successful CDI solution helps organizations in:

• Effective customer management by providing a timely and accurate understanding of customer needs and behaviors.
• Improved cross-selling and up-selling opportunities by understanding the prospective customers.
• Removing duplication and misleading customer information and providing single version of truth across the various business units of an organization.
• Providing effective campaign management.
• Complying with legislation, regulations and privacy requirements.
• Optimizing operational, maintenance and enhancement cost by having a central integrated environment (hardware, software).

Customer Data Integration - Challenges

In most organizations, since customer information is distributed across various applications, the unification and integration of customer information from heterogeneous and dispersed applications is a big challenge. Forester Research has found that though 92 percent companies say that having an integrated customer application is critical or important, only 2 percent have managed to achieve this. There are numerous challenges faced during customer data integration:
Duplicate Customer Data Duplicate customer records hinder the organization's ability to identify the customer uniquely and correctly. Duplicate records also cause problems in relating customer transactions to a single customer record. Also, it becomes difficult for the customer service representative to correctly understand the history of interactions made with a customer. The other significant drawback of duplicate records is that it causes duplicate campaigning. Key factors influencing data duplication issues are:

• Local maintenance and storages of customer information in an individual application;
• Inorganic growth of the organization (merger and acquisition) resulting in heterogeneous processes and systems to maintain and support customer information;
• Different customer details fed through different channels (Web, telephone, etc.);
• Data entry error;
• Relaxed data entry service level agreements (SLA) and audit;
• Lack of briefing, training and education to the customer service/front-end staff about the important and significance of customer data fields.

Individual systems have their own way of maintaining customer details, which lead to ambiguous and duplicate customer information. Data entry error or inconsistent data entry by customer service agents leads to an ambiguous data set. To compound the problem, the customer maintains different contact details when they are interacting through different channels. For example, the name and address can be captured in the various ways, which leads to duplicate or inconsistent customer details within and across the applications.
Inconsistent and Inaccurate Data Inconsistent and inaccurate customer data limits the organization's ability to understand and analyze the customer. This leads to poor decision-making that causes customer dissatisfaction. This inconsistent and inaccurate data set can generate a different version of the customer information and defeat the prime purpose of CDI, which is to produce a single version of the truth. It also leads to data reconciliation issues and affects the functioning of the business applications. Key factors influencing consistency and correctness issues are:

• Lack of common metadata control: Distributed and disintegrated customer metadata across application can lead to the inconsistent definition of the customer.
• Clerical errors (data entry error): Inaccurate and insufficient data entered by data entry operators or call center agents leads to data sufficiency and accuracy issues.
• Lack of data ownership, infrequent audit and relaxed SLA. Based upon business needs, individual business groups (sales, operations, marketing, human resource, etc.) primarily focus on the given subset of customer information. The data fields not being used by given business groups can contain a default or meaningless value in the database. Inconsistent domain range and or default values definition can generate the data consistency issues, i.e., default values for birthdate can be different for different departments. The missing data fields cause data sufficiency issues.

With the increasing volume and velocity of data, managing data growth and maintaining the latest and accurate customer information is a challenging task. Decayed and old data contains no value to the business. The two major areas of data management are growth and latest data management.

Growth management. Business applications generate millions of customer records every year. Inefficient data management and storage can have an adverse affect on the performance and usability of the application. Factors contributing data growth are:

• Nature of operational systems (business applications),
• Inappropriate historical data management strategy,
• Lack of data archival and housekeeping strategy, and
• Inappropriate reference data management strategy.

Current data management. Customer information changes over time. The CDI application should track the changes and maintain the most current customer information. Factor contributing information changes are:

• Changes in customer credentials;
• Changes in customer contact details; and
• Changes in customer demographic, psychographic and geographic details.

The CDI Solution Architecture

A customer data integration (CDI) system is the central application to capture, integrate and distribute customer information. The goal of a CDI application is to integrate customer information from different applications with minimum latency. Based on the needs of an organization and the dynamism of the customer data, the CDI architecture can be implemented either using batch processes (ETL - extract, transform and load) or using real-time messaging (EAI - enterprise application integration).
Figure 3: CDI Logical Architecture (Hub-and-Spoke Model)

A CDI system extracts customer information from disperse applications and performs data cleansing, customer matching (deduping) and integration as per the predefined cleansing, matching and integration rules. The central repository contains the integrated customer data with different views of customer information. The data access interface defines the data access mode, restriction and privileges. Business applications and user communities can access only that data set they are authorized to. Business rules (data cleansing, customer matching, data integration and data access rules) can be stored in the central metadata repository or reside in the individual tools repository.

Conceptual Data Model (CDM)
CDI is an application to store and distribute meaningful customer information. The CDI data model contains customer and related entities. The generic CDM is illustrated in the Figure 4.

Figure 4: CDI Conceptual Data Model

Customer and customer classification. A Customer is a Person or Organization of interest. Customers enter in a relationship with other customers. The nature of this involvement is used to determine whether a specific customer in an external customer, employee, supplier, partner or a competitor. The customer can be viewed from various perspectives as discussed in the beginning of this article.

Customer relationship. This entity stores relationship between two customers. Customer relationships can be categorized as personal or professional, e.g., Parent-Child, Employer-Employee, etc.

Customer contact. This entity captures the customer contact details. Customer contact details can be the physical address, telephone contact and electronic information. The postal address can be subgrouped as current address, permanent address, office address, bill-to-address and ship-to-address. Telephone contact consists of home phone number, office phone number, cell number, corporate office number and local office number. Electronic address consists of personal, office and corporate email ID.

Customer details. This entity captures the customer demographic, psychographic and geographic details. This information can be used for customer segmentation and analysis.

The demographic details to be captured are gender, age group, marital status, number of children, profession, income group, other financial details, etc. The psychographic information captured includes channel preference, privacy specifications, market research, etc. The geographic details to be captured are location (country, region), population groups, country development status, primary currency etc.

Customer accounts. A customer account is a contractual relationship between a customer and an organization and is associated with a given product or services. The account entity stores the account details, account type, account status and other related information.

Customer household and household details. Households are the collection of existing or prospect customers. Households and their demographic, psychographic and geographic information will help in understanding the associated patterns and defining the proactive campaign management.

CDI Processes
CDI processes facilitate the consolidation and unification of disparate customer data into integrated and meaningful customer information. The key driver for customer data integration is to provide the true view of customer. The process steps involved in the customer data integration are data acquisition, data cleansing, data integration and data management.


Figure 5: CDI Processes

Data Acquisition
The data acquisition phase helps in understanding the customer data and defining the data extraction strategy. It involves the identification, analysis and extraction of customer data from various business applications (operational systems). A detailed study of source data is performed to understand the data format, characteristics, pattern and usability. A data extraction strategy and approach is defined to extract the relevant customer information from source systems.

Data Cleansing
The data cleansing phase encompasses the processes and procedures for data correction and standardization. Data correction is the process of fixing, spelling and correcting the address, ZIP code, Social Security number and permanent account number. Once the data has been corrected, it needs to be standardized according to the predefined data format and structure through a data standardization process such as storing the Social Security number as 999-99-9999.

The data integration phase includes the processes for matching, merging and linking of customer information. This involves the following processes:

• Customer matching and linking - Customer data is deduped to remove the duplicate customer records and generate a single customer record valid across the business applications (source systems). Also, customer records get linked with the other related records, i.e., households and organizations.
• Data transformation and integration - Data will be transformed and integrated to produce the true view of customer. On a need basis, in-house customer information will be integrated with external third-party customer data set (e.g., Dun & Bradstreet, Experian) and produce the integrated customer database with various data access views.

Data Management
Data management includes the processes for monitoring and maintenance of customer data, which is dynamic by nature and changes over time. It requires periodic data monitoring and maintenance to keep the up-to-date customer information available.

Data monitoring processes periodically analyze customer data to understand any changes in the customer information. Data maintenance makes the latest information available and archives the old data set. The archived data set is required to reproduce the snapshot of customer information at any given point in time.