Log in

Navigating the Six Cs of Data Health

Is your data seaworthy? Is it clean, complete, and useful? This blog post explores the six Cs of data health which will help you chart a course toward more trust in your CRM data allowing you to make better decisions throughout your voyage.

Navigating the Six Cs of Data Health

Stormy Cs

You may be able to sail the seven seas, but when it comes to navigating data integrity, there are six Cs to consider. I’ve been working through Trailhead and other resources learning about Data 360, and the six Cs of data quality framework is a concept that shows up across a few different sources with slightly different names for each C (more in the Resources section below).

As someone who’s worked with CRM data for a long (long) time I’ve seen a LOT of data, especially since Arkus manages data migrations and health checks on a regular basis. While these Cs are already things Arkus keeps an eye out for, I appreciate how the six Cs organization is a clear and specific way to structure data health best practices. Whether you’re planning a data migration or you simply want to evaluate your current CRM’s data health, working through the six Cs is a useful way to approach your task. After all, your CRM is your system of record and a sound ship (i.e., data integrity) should always be considered business-critical, especially if you have multiple sources of data. The data may all wind up in the CRM, but it could be added via an online donation integration, marketing platform tool, or even manually imported from a siloed platform or spreadsheets.

Why does all of this matter? Because bad data leads to bad reporting and unreliable AI tool results, which throw you off course and set you up to make bad decisions. And remember, you and your organization’s users are not the only ones impacted by questionable data. AI tools that make use of that questionable Salesforce data are not going to be able to provide satisfactory results. Bad data can capsize your relationship with your constituents if, for example, a solicitation email sent to folks who have opted out. It erodes trust.

This blog post is intended to help you evaluate your data health by exploring the six Cs, why each matters, and what you can do to mitigate any gaps you find to right your CRM ship.

Note: There is a lot of overlap across the Cs. The Cs are heavily intertwined just as is your data.

Consistency

Consistency is the foundation of good data. Inconsistency is navigating without modern radar - you may reach the shore but you may have taken the long way around. Do you have data entry standards defined and do your users follow those standards? Data uniformity leads to full and accurate views of your constituents.

Why it matters:

  • Consistency builds confidence in your data and reporting both of which better inform decisions.
  • Inconsistent data can create headaches for departments (see the email example above)
  • Consistency makes it easier to prevent or, if needed, detect and resolve duplicates.

What to do:

  • Define standards and create a data dictionary to track all of your agreed-upon rules (e.g., US states should always be entered as the two-character state abbreviation - NY instead of New York).
  • Put checks and balances in place to allow for internal and external user error and conduct reviews periodically to ensure standards are being met (e.g., phone numbers entered that include text).
  • Ensure the right fields have the right field type. You shouldn’t enter an email address in a text field and vice versa.
  • Ensure the meanings of terms, such as those found in drop-down/picklist values, are consistent across records and across tables.
  • Implement validation rules to capture data for both immediate need and posterity (e.g., if a grant application is not awarded, users should note why it was denied in case you ever decide to apply again).
  • Pay attention to timestamps, such as the created and last modified dates to help with data cleanup down the road (e.g., merging constituents and knowing which has the most recent email address).
  • Add help text to fields and provide in-system procedural guidance for users wherever possible.

Completeness

Completeness is about ensuring information is filled in and taking action when key data is missing.

Why it matters:

  • Incomplete data can impact reporting, decision making, and marketing.
  • Missing data prevents you from having a full understanding of constituents and their activity.
  • Missing data may prevent automated processes from running.

What to do:

  • Define what data you need to collect so you meet specified use cases to ensure you have as full and accurate a picture of your constituents as possible.
    • Don’t forget the downstream effects! Not collecting a first name for a constituent is going to make email personalization difficult.
    • Been in your CRM for a while? See if your CRM allows the use of tools to report on field usage. That one checkbox that’s checked for only 5 out of 100K records probably isn’t terribly important.
  • Decide which data legitimately needs to be required. An email newsletter sign-up sheet requires an email address, first name, and last name, but a middle name isn’t as big a deal.
    • Be sure to take external data into account. You don’t want to cause your integration sync or manual upload to fail because you require a field the other platform does not or cannot require
  • Inspect and clean up your data regularly. Since you can’t require every single field, create reports and/or build automation to identify and fill in gaps or assign someone a task for follow up research (e.g., look for 111-111-1111 and 555-555-5555 phone numbers).
  • Provide training for new users as well as current users who need to improve their data entry practices. Be sure to emphasize that data integrity impacts everyone at the organization.
  • Track metrics on record completeness. Build a formula or automation that counts blank fields and weights them appropriately (e.g., a missing email address might be weighted higher than a missing home phone number). This allows you to prioritize cleanup and could even inspire an enrichment campaign asking email subscribers to confirm or update their contact information.

Correctness

Do you trust your data is correct? If you don’t have standard operating procedures for entering and validating data, by default the answer is no. When you can’t trust your data, you can’t trust reporting and analytics which means you can’t trust the decisions you need to make based on the reporting and analytics. Correctness leads to truth and trust.

Why it matters:

  • Invalid contact info like junk phone numbers and junk email addresses means poor communication with your constituents
  • Bad data leads to bad reporting and bad decisions.
  • Agents and other AI tooling referencing incorrect data can create cascading errors

What to do:

  • Validate data at the time and point of entry to detect inaccuracies and junk data using simple reports or robust analytics tools.
  • Rate fields by correctness
  • Add validation rules to ensure data is being entered correctly
  • Manage duplicates regularly

Current

While your CRM will experience constant activity, there’s always going to be a subset of data that gets stale quickly. Sailing with an old chart might get you close to where you want to go, but shifting sands may cause you to run aground. Consider this scenario…

You want to migrate historical, manually uploaded wealth screening data from your legacy CRM to Salesforce. However, your project scope also includes integrating a wealth screening tool. With real-time wealth data being synced to Salesforce, is the old static wealth data really needed?

Why it matters:

  • Stale data can skew report results which can lead to improper conclusions and/or bad decision making
  • Stale data could be anywhere in your CRM. Outdated mailing addresses lead to lost event invitations and missed chances for RSVPs.

What to do:

  • Define which data you believe to have an expiration date and review it regularly to ensure it reflects the most recent understanding. Stale data isn’t necessarily bad, but you need to be able to recognize it.
  • While defining (and cleaning up) dormant data, use it to determine patterns. When does the data go stale and why? For example, work emails, phone numbers, and addresses will go stale shortly after a change in employment.
  • Be sure you’re able to exclude stale data from reports.

Contextual

Managing data is about quality over quantity and quality means context. Overloaded ships have a tendency to sink. Ensure you’re collecting data that serves your business requirements. Everything else just weighs you down.

It should be noted that a field or piece of data can meet the requirements of the other Cs, but that doesn’t mean it belongs in your CRM. “Favorite Color” may be entered consistently and correctly, but does knowing that really matter to your business?

Why it matters:

  • Irrelevant data can confuse users and, as such, is less likely to be used. Does that field need to be filled in or not? When people don’t understand the downstream effects of skipping entering data in a field, the lack of context can hinder collaboration down the road.
  • Extraneous fields can overcrowd a page layout making users less equipped to enter the right data in the right places or discouraging them from trying to enter the data at all.

What to do:

  • Create a data dictionary with guidelines that explains what information needs to be captured and why.
  • Audit field usage regularly. As touched on earlier, that one checkbox checked for only 5 out of 100K records probably isn’t useful.
  • Use descriptive field names (e.g., “Lifetime Value in $K”) and include help text if possible (e.g., “Lifetime Value in Thousands of USD inclusive of hard and soft credits”)
  • Carefully consider use cases before adding new fields. Users are going to ask for new fields all the time, but take the time to consider if that new field just serves the needs of only one or two individuals or the organization as a whole. There may be better and/or out-of-the-box ways to capture the requested information.

Compliant

Compliance is relevant for your data whether you’re sailing in international waters or traversing a local river. Obviously, key private information, such as social security numbers, driver’s license information, etc. must be saved properly and only be seen by certain users. And it’s clear that any medical related data must be HIPAA compliant in the US. But compliance doesn’t stop there.

Why it matters:

  • You don’t want to get in trouble with the Federal Trade Commission (US) or European Union. CAN-SPAM and GDPR both require you to follow certain rules regarding email marketing. Not following the rules can leave you open to financial and legal trouble.
  • Data that’s out of compliance risks your reputation as an organization, which leads to a loss of trust. (And see the previous point about potential financial and legal trouble.)
  • Setting up your CRM from the beginning to adhere to compliance is much less effort than responding to a possible breach, which would take both time and resources.

What to do:

  • Confirm you have consent to collect data and clearly define the levels of consent (e.g., is your constituent opting into all emails or just education-related emails?)
  • Understand the provenance of your data. Where did it come from (e.g., through an online donation platform or an email sign-up), when was it received, and when was it updated? The date and time a constituent record was created isn’t necessarily the actual date and time the information was collected.
  • Ensure you protect, encrypt, mask, etc. your data in accordance with the law and best practices.
  • Control access to data through permissions and record sharing rules.

Plan to Keep Your Data Shipshape

The primary focus during this discussion has been related to your CRM, but constituent data could be stored anywhere. You may have multiple instances of Salesforce or an external email marketing, volunteer, or event management platform. (You’re 100% on the wrong course if you’re using spreadsheets to manage constituent data.) Each of those platforms should follow the six Cs. Why? Because inconsistent, incomplete, incorrect, out of date, out of context, and out of compliance data creates navigational hazards and puts your ship at risk.

That brings me back to learning about Data 360. Data 360 is a powerful toolbox that has the ability to harmonize, segment, and analyze your data from multiple data sources (not just from your Salesforce CRM) to provide insights on your constituents as a whole. Not just donation history or purchase history or volunteer activity in isolation, but a true 360° view allowing you to make decisions, use your data history to predict trends, and deliver personalized experiences to your constituents.

Nearly every action you take in Data 360, from ingesting, to harmonizing, to activating consumes credits, and credit consumption is based on the volume of data you’re processing. You have the power to keep credit consumption costs down by adhering to the six Cs guidelines above. Regardless of whether you’re planning to use Data 360, following the six Cs will provide enormous benefits. You’ll be able to trust your data, empower your users, and make better and faster decisions.

Want to talk more? Reach out to Arkus and ask how we can help swab the decks and make your data more seaworthy.

Resources

Note that Cs may have slightly different names across resources, but the concept is the same.

This blog post was written by the author. An AI tool was used for copy editing.

BlogData StrategyGovernanceBest PracticesData 360CRMSalesforceAdmin TipsNonprofit