Data decays at an astonishing rate. Companies relocate, hire and fire, and change constantly. A data point verified last week is likely already out of date. For better and worse, the data that is most valuable for sales and marketing teams is inherently dynamic.
For the first time, high-value, rapidly changing company insights are readily available, but keeping up with that data is challenging.
Why is it important to clean your CRM?
Any time you reach out to a prospect with outdated or incorrect information, you are already behind your competition. A CRM overrun with duplicate and outdated account information can have major implications - from wasted prospecting time on duplicative efforts, to countless missed opportunities.
Before you can implement any of the tools that help you keep up with rapidly changing information, you must make sure that your CRM is thoroughly cleaned and primed for future enrichment and scalable prioritization. With a clean CRM, organizations can set realistic expectations for the amount of data enrichment a data provider can offer. The first step in this process is to assess your current data quality.
Diagnosing data anomalies in your CRM
A CRM diagnostic can be done internally, or performed by a data partner that offers one. A comprehensive diagnosis will pinpoint where data quality is breaking down and give you an idea about the scale of your data integrity issues.
Without a good grasp of your CRM platform, you can't treat the origin of the integrity problems and accurately evaluate data providers:
Understanding your CRMs' data anomalies is the first step to mitigating them in the future: a CRM swimming in duplicates might require more advanced duplicate identification rules, while inconsistent data format could point to a need for standardizing data entry.
If your CRM database is overrun with dirty data, it will be impossible to tell if the provider has poor matching logic, poor coverage or if it's your data. Dirty data will never match well to an external data set. If your CRM is overrun with dirty data, any evaluation of data vendors will return poor matching results --- even the best algorithms cannot match to dirty records.
Data quality is determined by the prevalence of data anomalies. The type and rate of decay varies from organization to organization, but the most common types of data anomalies are duplicate, missing and dirty data.
- Key identifiers are in conflict:\ (e.g. Uber's account URL is "www.lyc.com")\
- Key data is outdated (e.g. Gusto's account is s1ll named "ZenPayroll")\
- Data points for the same field are not standardized across records (e.g. email domain is in a URL field)
- Multiple records for the same account
- A substantial portion of records have null values for one or more important fields
Understanding what causes data anomalies in the first place
Data can decay in infinite ways - naming conventions vary between data providers, errors in data entry can pile up, and company information can become outdated. Over time, this results in a CRM overrun with multiple records of the same account, empty fields on accounts, or unreconcilable information within the accounts. Understanding how where and how the anomalies are occurring will help you fix and prevent them.
Duplicates are almost a given when managing multiple data inputs (including humans), but implementing stewardship rules in your organization to standardize data inputs can help. Irreconcilable data points can be helped by limiting access to the CRM.
Amending the data anomalies
While a data assessment can be done manually, working with a provider that has a diagnostic solution will save a lot of time. With a provider, matching is typically phase one of the deduplication process. Powerful matching can amend account information despite inconsistent conventions -- ensuring that a high percentage of duplicate accounts are captured, rather than lumped in with dirty profiles. Matching helps identify duplicates, but no matching process can amend dirty accounts (because they can't be matched in the first place).
Equipped with a comprehensive assessment, you are now ready to begin the process of cleaning your CRM. For each type of decay, the solution can vary. Typically, organizations merge duplicate profiles, remove dirty accounts that cannot be rectified, and enrich empty fields with account data from an external provider. For details on how to nail the enrichment process, check out our post on matching and enrichment.
The clean-up process for dirty accounts must be performed manually, because dirty internal data cannot be amended by a provider. Amending dirty accounts requires input from the account owner to determine whether an account named "Uber" with a URL of "email@example.com" should be deleted or cleaned up.
After identifying duplicate profiles in the diagnosis, the profiles must be merged into one. Salesforce offers a number of easy deduplication apps (like RingLead) that can can perform this automatically. Microsoft Excel or Google Sheets are standard for manual deduplication.