CRM Software Quotes
CRM search»CRM Implementations»Lessons Learned in CRM Data Conversions

Chuck Schaeffer Lessons Learned in CRM Data Conversions

4 stars Average rating: 4 (from 124 votes)
By Chuck Schaeffer

Customer Data is Your Most Valuable Information Asset - If Attended To

One of the great potentials of Customer Relationship Management (CRM) is derived from the actionable information and insight that can advance customer relationships and the company's business model. CRM software can compile data in order to deliver intelligent information to knowledge workers and decision makers such as the factors that influence customer relationships, key performance metrics, sales forecasting and pro forma business modeling. However, data quality must be in tact for information to be believed, trusted and acted upon. This is why data integrity and data quality measures are crucial to successful CRM strategies and software implementations.

Despite lessons learned and best practices regarding data conversions, many CRM implementations incur an early project delay because of a failure to survey the data prior to the conversion - followed by the untimely discovery of dirty data, duplicate data, incomplete data and bogus data. This often results in added tasks to the project plan for data cleansing processes and puts the project team behind plan early in the project.

Data Conversion Planning and Cleansing

Data quality surveying and planning measures must be initiated at the beginning of the CRM implementation project to mitigate project delays. During the implementation of the system, legacy data is usually electronically converted into the new CRM system. This data needs to be reviewed, scrubbed, filtered and possibly appended before it is migrated to the new customer management system. A systemic process must be implemented to ensure valuable data is not lost and bogus data is not transferred.

First, map your old CRM system to new CRM system data fields. If there isn't a field in the new CRM, decide if the legacy data is something that is useful. If so, does your new CRM system give you the flexibility to create user-defined fields? If not, software customization may be needed. Alternatively, consider mapping to other unused fields in the new CRM application.

When mapping the data fields, make sure the field format of the target fields are the same and the size will accommodate the maximum length of your existing data. If the target field is smaller and there is no way to increase the field size, consider ways to abbreviate your legacy data or truncate the data.

Take the time to manually review the data and verify the age, utilization and value of the data. Are there customers in your legacy data that have not done business with your company in a very long time? Are there contract or warrantee considerations to keep data? Are there legal issues? Remember, you can archive your data to satisfy regulatory concerns rather than convert the data to a new system if it's not going to be used.

Converting bogus data reduces overall data quality and trust in the new system. Before importing previously purchased lists, look at the age of the lists. Contacts on lists become inaccurate at a rate of 2% per month. If you have a marketing list that is three years or older, you may be wise to either clean it up or discard it rather than convert it to the new system. Even contacts known at one time come and go. Make sure you are not importing bogus contacts. Most contacts should have some relevance, someone you have talked to and done business with, not just someone you emailed three years ago and have never heard from. Do these contacts have addresses, phone numbers or emails? There is often no point in keeping just contact names which may or may not be valid.

Customer Data Verification

The best way to ensure your customer data is correct is to verify it. There are third party services that can take your company and contact lists and verify the data for you. These services typically have databases that your data can be verified against. These services may then call and verify the contacts that don't match up.

Better yet, months before the CRM implementation project starts, have your customer service add contact verification procedures to their normal support or outreach processes. At the end of each customer service call, your customer representative can ask the customer to verify their information. This may only take care of your current customers however these customers are likely the most valuable and further, if you later elect to use a third party verification service the fees will be reduced by the number of contacts previously confirmed.

CRM Data Standards

Once the customer relationship management system is in production, there are several procedural controls you can implement to maintain data quality. First, institute a data quality policy that includes data entry standards and procedures. Data standards should address whether your data should be in all caps or mixed caps, standard abbreviations, minimally required data, data formats such as phone number formats and date formats. Eliminate spelling errors by creating pick lists for some types of data such as countries and states. Consider other cases where you can use a pick list (which should be small enough that users can use with minimal or no scrolling). Such cases may include industries, revenue ranges, products, etc…

It is a good idea to develop a customer naming convention so that companies are consistently named, thereby ensuring common understanding and quick retrieval of customers as well as reducing the likelihood that the same company will be entered under two different names. Below are some customer account naming best practice data quality measures to consider during your implementation:

  • Multiple Names: In order to ensure consistency among customer naming, you may first want to determine which name to use for customers that are referred to by more than one name. For instance, determine whether your policy is to enter the customer's legal name or the customer's most recognized name. As an example, determine whether your policy is to enter "Sears, Robuck & Company" or simply enter "Sears".
  • Acronym Names: Determine whether or not, or under what conditions, to use or not use acronyms as customer names. For example, is the organization entered as "America Online" or "AOL".
  • Name Prefixes: Determine whether it is your policy to maintain or remove a prefix, noun, pronoun or similar descriptor preceding a customer's name. Alternatively, just decide whether to include or remove the word "The" when it proceeds a customer's name. For example, do you keep or remove the word "The" when entering the organizational name of "The County of Palm Beach", "The Titan Companies", "The Hartford Group" or "The New York Symphony".
  • Abbreviations: Determine whether you prefer to use or avoid abbreviations with customer names. For instance is the organization entered "US Steel", "U.S. Steel" or "United States Steel". If you elect to use abbreviations with customer names, determine whether the abbreviations should be entered with or without periods.
  • Avoid using these special characters in the customer name: \ / : * ? < > | # ", $, &, [ ].

Establish minimum data requirements for various types of records or transactions. Names and phone numbers may be sufficient in some cases. However, if you are a company that relies heavily on email marketing, an email address should be obtained. Many customers even prefer to be contacted via email rather than phone. Gender and age are tricky but you can provide a range of ages to which customers are more likely to respond. Other useful information such as household income or company revenue (for B2B customers), date of last internet purchase, or even which department stores customers buy from can add data intelligence for later target marketing, product promotions or loyalty programs.

To fill data gaps, you can also charge your customer service reps to collect the needed information at every opportunity. You can make it relatively painless by adding a few lines to call scripts. During every call, you customer rep can ask the caller to validate the information most important to your business. In addition to contact information, don't forget about data such as job titles, which can be revealing information, suggesting whether the contact is a decision maker or influencer, and leveraged for highly specific marketing campaigns.

Data De-Duplication

Most CRM systems suffer from duplicate data and the longer the problem goes unaddressed the more the risk to data integrity and loss of user confidence. Even when users are taught to perform a search before entering a new record, in the heat of a customer support call or a new hot opportunity, users sometimes fail to first verify the new record is not already in the system. The three measures required to combat this common problem are management enforcement of data quality procedures (i.e. identify violators and take corrective action), implementation of automated data checks at the source of new entries and periodic data duplication checks with follow-on merging of duplicate records.

Several CRM systems offer programmatic routines such real-time data duplication notifications that warn you if you are about to enter a duplicate customer or contact. In some cases, duplication may be acceptable pursuant to your data quality policies, such the same company in different locations. Nonetheless, a warning or an option to merge the new information with an existing record can be helpful.

Most CRM systems have batch oriented duplication detection and merging capabilities. However, someone needs to be charged with initiating the process, reviewing the recommended reports, determining the data that should be deduped or remain segregated, and then for records to be merged, determine which record is the master and which is the subordinate.

There are also third party data quality vendors which integrate to the customer records at the time of entry or edit, in order to validate names and addresses against the U.S. postal service or other databases, and make both correcting entries as well as populate missing fields. These services can automatically insert missing postal codes based on the street address, or vice versa, correct telephone area codes, enforce common formatting characters such as brackets or dashes on the telephone number and even apply rules to identify that IBM is the same as International Business Systems and insert the correct name pursuant to your data quality procedure.

No CRM system can remain clean without a data guardian, someone who is responsible for ensuring data quality standards are being followed and periodically sweeping the database for duplication, errors and relevance. Many large companies have teams of people whose responsibility is to ensure data integrity.

Global Data Considerations

For international companies, data quality takes on more challenges. The database may have to be configured to support multiple languages. Multiple dictionaries may have to be added for spell check utilities which operate standalone or in conjunction with automated duplication utilities. Non-standard or double byte characters such as Chinese, Japanese or Korean may not work with several CRM systems data deduplication routines. If your CRM software application supports both standard and non-standard characters, it is best to separate the non-standard characters to different localized fields. Many companies require that primary key fields are maintained in one language, usually English, while localized fields can be used for other languages. This will enable global users to be able to search in a standard language.

Dates and currencies may require advance planning in global CRM implementations. More than just the formatting of dates, how will your reports work among different time zones. Something as simple as a monthly sales forecast report can prove frustrating if data structure isn't established early. If your sales manager runs her monthly report, does it include opportunities from midnight of the first of the month to midnight of the end the month? Whose midnight? The opportunity owner or the sales manager? If her VP in London runs the report, will the two reports be different? A global or standard time zone for reports can resolve many of these issues.

Currency issues come in a variety of scenarios. How often will your system convert exchange rates? Will the exchange rate type be a buy rate, sell rate, spot rate or something different? What currencies do you use for reporting? If you run a report today for last month, does the report reflect last month's currency exchange rate or this months exchange rate? Again, having some policies and procedures in place will alleviate these issues. Global companies have an official 'base' currency, usually dependant on the company's headquarter location. Decide whether reports run on the current rate or retro to date of transactions - one or the other, not both. If you create reports with both options, make sure your users understand the output. Trending gets a little tricky but if you establish the rules, your trending assumptions should include these policies.

CRM systems require care and feeding. Managing them is like taking care of a living organism; you must feed it, establish boundaries, and keep it on a regular maintenance schedule. Your customer data is quite possibly your most valuable information asset. Your CRM system should enable you to store, protect and benefit from this asset, such as delivering the information to better understand your customer relationships and how to better serve them. End

How would you rate this article?   

CRM Software Category
 Filed In Categories: CRM Software Implementations
User Adoption Tag
 Tags Tags: Data Conversion
Trackback
 Trackback Permalink: www.crmsearch.com/data-conversion.php
Author
Author  Author: Chuck Schaeffer
Share
 Share Share:    Bookmark and Share
CRM

 

Comments (2) — Comments for this page are closed —

Guest Craig Tulley
  Can you recommend one of the data quality services which validate proper company addresses?
  Chuck Chuck Schaeffer
    In my prior CRM software I used the address verification service from Melissa Data. As the CRM system was online, and Melissa Data also offers an online service, the integration was fairly simple. That's the only service I have first hand experience with so I recommend a Google search to uncover others.
 

 

Share This Article

 

Quote

Despite lessons learned and best practices regarding data conversions, many CRM implementations incur an early project delay because of a failure to survey the data prior to the conversion - followed by the untimely discovery of dirty data, duplicate data, incomplete data and bogus data. This often results in added tasks to the project plan for data cleansing processes and puts the project team behind plan early in the project.

 

Related Articles

 

More Articles By Chuck

 

 

CRM Price Quote

Follow Us
social
social
social
social

crm search

Home   |  CRM  |  Sales  |  Marketing  |  Service  |  Call Centers  |  Channels  |  Resources  |  Blog