Kris Bochat, Senior Consultant, LexisNexis:
“A-B-C, easy as 1-2-3” … at least that’s what the Jackson 5 used to say.
Greetings, fellow CRM owners. In today’s CRM environment, we see a lot of clutter within our databases. We have contacts that date back to the first day we started, or we have never talked to, or now have no way of talking to. In my last blog post, I spoke to “saying goodbye to an old friend,” the premise of which was getting rid of those contacts we just no longer need no matter how painful it is to say goodbye. Now that you have put in the hard work and “trimmed the fat” in your CRM, it’s time to put together a plan with a few A-B-C steps to follow to more easily maintain the data you have left.
Today I will take you on a journey through our data maintenance plans that we commonly offer as part of our consulting services engagements. Although I can’t put a full plan on here because they can be more than 13 pages long, I can give you the basics to put together your own plan to get you started. Here is a basic outline for you to put together your own plan to help you get started. Easy as 1-2-3.
Data quality management can be roughly split between objective and subjective tasks. In some cases, these tasks may be assigned to different data stewards based on experience and preference. In other cases, both types of data cleanup may be performed by the same data steward(s), and the difference may relate more to scheduling of certain cleanup tasks.
Objective DQM refers to data quality issues that may be handled and managed by a data steward without accessing the Data Change Management (DCM) inbox or, generally, contacting end users to verify data. It generally requires less familiarity with the collection of contacts being referenced and can be managed using bulk data-cleanup searches and tools in the Windows Client.
Objective DQM tasks include:
When reviewing a new contact:
After reviewing a contact and standardizing the data, remove the contact from the New Contact Review folder.
SCHEDULE: New contact review processes should be executed daily.
Duplicate contacts can be identified in InterAction® using the potential duplicate searches, Prioritized Data Management Queries (PDMQ), or by reviewing new and possible duplicate contacts. Duplicate contacts can be merged using the side-by-side duplicate merge functionality or by using the Multi-Contact Duplicate Merge (MCDM) utility. Duplicate contacts should never be resolved by simply deleting one of the contacts from the database. Finally, the “loser” contact(s) are moved into the Duplicates folder after a duplicate merge has been completed. The contacts in this folder should be deleted on a regular basis.
The firm should begin to manage duplicate contacts by comparing new contacts to the firm list to confirm a duplicate contact does not exist, and by using MCDM and the Potential Duplicate Contact searches to identify and merge duplicate contacts across folders and searches. MCDM allows contacts to be grouped by a variety of criteria; data stewards responsible for ongoing duplicate management should process the MCDM results by the most restrictive criteria first. Generally, company duplicate merges should be performed before person duplicate merges; one of the criteria for person merges is the company name, and performing the company merges first will standardize that information.
SCHEDULE: Duplicate contact management tasks should be executed on a rolling schedule. If available, manage high-priority contact collections first using PDMQ, and then work on the general firm list.
In an actively used CRM database, marketing lists used for mailings, events, etc. can quickly overwhelm users. List management is an easy task that anyone with the proper access can manage easily, and in most firms, the person in charge of the list is the one who will manage it to the end. Most firms will have events where a set of contacts are invited to a gathering where RSVPs and attendance are tracked. Following the event, the list is reviewed, and the appropriate information is disseminated to partners at the firm for follow-up. If an event, mailing, or other type of marketing activity occurs once a year or less, the list should be archived to clear up space for new marketing activities.
SCHEDULE: Create a temporary archive folder type under your marketing lists to move events after they have occurred. After a year, or after you have created a new list from an old one, archive the list into the administrative archive folder. Finally, after two to three years delete lists.
Subjective DQM generally requires a judgment be made as the quality and accuracy of the proposed data change is approved or rejected. Data stewards with subjective data quality management responsibilities spend more time in the DCM inbox and a larger percentage of their time doing research, contacting users to verify changes or obtain additional information. In the past, researching information about contacts required a lot of additional time, which sometimes led to dead ends. With products such as InterAction IQ, getting contact information has never been easier. Managing data quality and prioritizing new contact information becomes even easier when you can pick IQ data over older, user-contributed data.
Subjective data quality management tasks include:
Conflict tickets are generated during a first-time User to Firm Sync with what InterAction considers “suspect contacts”—contacts where InterAction cannot determine the last edit date of the contact, such as those brought into InterAction from Outlook for the first time, or those created when contacts are imported directly into a user contact list. They are particularly common when new users are brought onto the system. When InterAction recognizes that a suspect user contact and a firm contact should be linked, but the user contact contains conflicting information, a Conflict ticket is created. These tickets will be generated in waves and should generally receive a high priority. Evaluating Conflict tickets requires that the data steward work with the user who generated the tickets to understand the relative accuracy of the user’s data in Outlook. All Conflict tickets are Submit tickets.
SCHEDULE: After onboarding a new user.
Edit tickets are generated when contact data is modified in the web client or in Outlook. These change tickets should generally be prioritized by contact type using the comprehensive view in DCM inbox; contacts with the oldest tickets should be reviewed to ensure that the data in InterAction is up to date. Generally, Submit tickets should take priority over Review tickets, but the data stewards should continue to review all tickets relating to a given contact at the same time (by following the “[X] open tickets for [contact A]” link.)
SCHEDULE: Ongoing. Use the DCM by Contact View to prioritize your firm’s most important contacts first.
User request tickets are generated when a user clicks the Change Request option on a contact in the web client. The change request wizard prompts users to choose from a pre-set list of change types, or to choose Other, then provides a space for a typed description of what the user would like to accomplish.
User request tickets are also generated when a user merges their own contacts in the My Contacts view.
User request tickets should receive a high priority, as they are generally created manually by a user.
Here is a sample plan that we put together for clients using the different data quality approaches:
My closing thoughts are to keep it simple. Put a plan together to manage your data, keeping it from becoming overwhelming. Use the data quality tools provided to make your life easier. You have worked hard to get your CRM database to where it is—keep up the good work!
P.S. As of the writing of this blog post, we have released new functionality in our Data Minder tool, which is one of my favorite new tools over the last several InterAction releases. The duplicate merge screen, machine learning, data normalization, and new contact health module will blow your mind, and I cannot wait to take our data quality to the next level. Stay tuned!