All Things Data

Customer Data Integration Experts to Speak at Insurance Direct Marketing Forum Session

I wanted to take this opportunity to announce that Bob Bloom, President of Business Development Solutions, and our client, Tim Casey from Conner Strong, will jointly present a case study on utilizing customer data integration to develop a single customer view for sales and marketing analysis and action.

The session is part of The 6th Annual Insurance Direct Marketing Forum in Philadelphia on September 15, 2008.  

So if you are going to be attending the conference, please drop by and say hi to Bob.

To view the press release:  http://www.prweb.com/releases/2008/08/prweb1212374.htm

External Business, Institutional, and Specialized Marketing Data, Customer Data Integration, and a Single Customer View

Many organizations are using external business, institutional, and specialized marketing data to enhance their single customer view.  However, there are many different data providers and the question is which one to choose.

If an organization has a relatively broad market, the answer may be that no single data vendor can meet all of the objectives.  For example, an organization may sell to a variety of for profit organizations, non-profit organizations, and government markets.  It might take at least 3 different data source providers to develop a single customer view that includes external marketing data for each of these different markets.  Further, even within each of these general markets, there may be sub-markets that require different external sources.  Public universities and federal government agencies both broadly fit under the category of government institutions and yet the best lists for these groups may come from completely different companies.

In addition to choosing external data based on the market, additional in-depth information might be needed to be included in a single view of your customers to help up-sell and cross-sell existing customers or prospect to potential customers.  A general compiled business list may be a good start to find basic information about a customer or prospect such as the name of an organization, the location, a general description of the business, the CEO, and other basic information.  However, if other senior level personnel in the business or more detailed information about that organization are needed, additional lists may be required to build the database.  There may be specific key indicators of an opportunity that are contained in Public Record files such as UCC filings, ERISA, or Workers’ Compensation information or there may be certain business intelligence that has been researched, such as what type or how much equipment an organization owns, by a different list provider.

The challenge in utilizing all of this external marketing data is keeping it properly organized.  Further, there most likely will be overlap and duplication between some of the external files.  A Customer Data Integration (“CDI”) application can keep this information together and correlated.  A CDI application can enable the best business intelligence from each file without the perils overlap and duplication.  By appending Common IDs to records that are part of the same organization or are related to an organization, these multiple files can be correlated to records in order to build the appropriate single view of each customer and prospect.

Sales can have a single customer view that is based on both internal and external data in order to help cross-sell and up-sell customers and prospects.  Marketing can analyze data based on a combination of internal data from disparate data sources and external business intelligence.  Further, the perils over counting the same organization as more than one “opportunity” during data analysis can be reduced since the multiple representations of the same organization will have the same Common ID as part of the Customer Data Integration process and therefore will only be counted as one.   

Customer Data Integration and Project Management

Like any information technology project, in order for a customer data integration project (CDI) to be successful, someone needs to take ownership of the initiative.

Therefore, one of the first items to consider when developing a single customer view is designating an individual who is going to be in charge.  This person needs to be available to discuss issues with the outside firm who shall be supplying the CDI software and services, coordinate internal resources, and make decisions where appropriate.

The following is an example of the type of decision that needs to be made in a customer data integration project to build a single view of the customer:

A group of records may be identified as being part of the same organization in an initial CDI match.  Within this cluster match, a decision could be made to identify those records that are at the same location get the same Site ID, those records that are part of the same Subsidiary/Division/Department get the same Account ID, and those records that are part of the same overall Organization/Corporation get the same Corporate ID. 

The rules for the customer data integration match, the number of corporate hierarchy levels, what the levels mean, and other considerations will vary client by client depending on what is needed for the single customer view.

Customer Data Integration Project Phases

Today, I thought it would be helpful if we took a look at the steps that are needed in order to have a successful Customer Data Integration (“CDI”) implementation.  We see the typical phases for most projects as Discovery, Design, Prototype, and Implementation.

The Discovery is the initial phase where the client’s goals are discussed.  Here, the client provides extracts of the data sources that are going to be used to develop the single view of the customer.  Once the extracts are received, we begin the process of reviewing the data sources so that we can understand the data in detail and note our observations in order to develop a strategy to get the best match results.  As part of this process, we document the data elements of each data source so that we have a resource that we can refer to during the other phases.   

The Design Phase is where the decisions start to take place.  Based on our initial data review and observations from the Discovery, a test match is set-up in our environment.  Initial match rules are set and there is a back and forth process with the client in order to finalize the match results.  Additionally, decisions are being made as to the various levels of the corporate hierarchy.  Other decisions regarding the workflow, security rights, data pre-processing, and other consideration are also being made.

The next step is developing a Prototype.  If the “go live” system is going to be hosted at the client site, the CDI software needs to be set-up in the client’s environment.  Regardless of where the database is hosted, testing needs to be done regarding the access to the client’s data sources, running initial and updated match results, reviewing and resolving results, and how the client is going to use the match results.  No matter how much Discovery and Design planning that you do, there are always details that cannot be uncovered until you actually start using the system.  It’s always best to make adjustments based on these findings in a test environment before the system is rolled-out.  Depending on the types of users of the system, some of the users may also be trained during the Prototype Phase.

The final step is to implement the single customer view in the “go live” environment.  Users that have not already been trained are trained on the specific aspects that they are going to be use.  The database administrator needs to monitor and maintain the application on an on-going basis. 

It should also be noted that although there is a phased approach to a successful customer data integration system, there is also overlap among the phases.  Typically, we do an initial test match, design, and prototyping very early in the Discovery to help us with our initial observations.  Further, it is helpful for the client to make Design decisions if they have a starting point.  Discoveries are always being made at every point throughout the process.  Adjustments and additional benefits and functionality can be implemented at any time, even after the initial “go live” system has been installed.

Customer Data Integration and CASS Software

As an organization, we are always looking for any advantage that we can get when it comes to matching as part of a Customer Data Integration project.  We try to use any piece of data that we can in order to develop someone’s single customer view.

Occasionally, we run into a situation where the client does not have sufficient data on a significant amount of records, such as missing zip codes.  One possible way to add missing data elements and help the match results is to use CASS (Coding Accuracy Support System) software.

According to Wikipedia, CASS software will correct and standardize addresses. It will also add missing address information, such as zip codes, cities, and states, to ensure the address is complete.  CASS software is offered to help organizations standardize data in order to gain postal discounts.  However, we have used it as another tool in our arsenal to help add missing data elements in order to develop a complete single view of the customer.

A Data Filtration System

I can’t tell you how many times when our company starts talking with a client about a data integration or other data management project that before we get started with the main purpose of the project the client first thinks that they have to “clean-up” the data.

If your organization has multiple data silos and has a significant amount of data records, a data clean-up effort could be a massive project.  Most likely, your organization will never get around to cleaning up the data, and if does, its going to be a one-time event that only provides clean data for a moment in time.  Therefore, if you are waiting for the day that you have perfectly clean data before solving your other data issues, that day will never come.

Instead, we suggest that your organization develops a Data Filtration System.  A Data Filtration system is a process where you can clean-up data on an on-going and systematic basis.  The system should match and link same and related organization and contact information from disparate data sources as well as within single data sources so that you can have a single view of your customers, prospects, and other organizations. 

With this single customer view, the data should be sufficient so that a person can make a determination on how to handle a group of records that have been matched and linked by your numbering schema.  Sometimes, a best record can be made from the group and the remaining duplicates can be purged.  Other times, however, records that at first appear to be duplicates are in fact records from the same organization but are separate sites.  Instead of merging and purging these records, these records should be kept separate but correlated by your numbering system.  There are other times when some records in your group need to be merged and purged while other records needs to be correlated.  Whatever the scenario, the system should be set-up so that these “Solomon” like decisions can be made.

Welcome!

Welcome to our blog, All Things Data.  I am Steve Kahn, Vice President for Business Development Solutions ("BDS").  BDS is the developer of SingleVision®, the data integration software application that integrates organization and contact information from separate data silos.

Here, I will discuss all things related to data integration, data quality, data consolidation, data hierarchy management, and other data management issues.

So why Blog about Customer Data Integration and related data issues?

According to the literature from the CDI ("Customer Data Integration") - MDM ("Master Data Management") Fall 2007 Summit that was held in New York, the industry trends are as follows:

• By 2008, the market for CDI solutions (software and services) will exceed US $1B   
• Additionally, Gartner confirmed the importance of this emerging market by issuing their magic quadrant on CDI hubs during summer 2005.  As did Forrester with their CDI Hub WAVE assessment.
• IDC has projected the overall MDM market to exceed US $10B by 2009. 

Still, while the CDI & MDM markets are fast growing, there are not alot of resources providing insights into these emerging IT segments.  My goal is that you will find this forum to be a valuable resource as you tackle your customer data integration, data quality, data consolidation, and data hierarchy management issues.