Well-designed data architecture provides “leverage points” for customer relationship management (CRM) and its connected systems. Organizations may obsess over a desire to use artificial intelligence (AI) features, automation, and predictive analytics, but the quality of your output is a direct result of how well these tools can leverage your data. This starts with data quality and data design.
Challenges With Data Quality
Every organization struggles with some level of data quality. One common challenge is duplicate records. Dynamics 365 provides duplicate detection rules to help find and root out duplicates. These alerts and merge functions help users handle duplicates, but even merging itself can skew reporting data (such as choosing which Lead Source data to keep).
Other organizations struggle with data inconsistencies: important fields are not filled out or are filled out inconsistently. These scenarios can be caused by too many required fields, not enough required fields, poorly designed option sets, or cluttered forms/views.
Some environments allow users to create records with little to no meaningful information, while others require so much data upfront that users avoid entering records altogether. Both of these approaches can reduce the system’s usability and trustworthiness.
Relevant Data Points at the Right Stage
A more effective approach requires information progressively as the customer moves through the business process flow. Instead of demanding every detail when a lead is created, organizations can require relevant data points at the right stage.
For example, a lead in a “Marketing Nurture” phase may require only a name, email address, company name, country, and marketing consent. As the lead progresses, more details such as Rating, Timeframe, and Estimated Revenue are collected. This approach improves adoption while still supporting data quality. This is not a prescription. A field like “Rating” or “Timeframe” does not always add value to the data set. If these fields are not actually being used for filtering views, prioritizing follow-up, reporting, or pipeline forecasting, they may not be worth requiring. Just because a field exists does not mean you should use it.
Excessive Customization
The opposing side of this issue is excessive customization. Every CRM needs to be tailored to the organization using it. However, because Dynamics 365 makes it easy to create fields and change option sets, organizations often introduce unnecessary complexity for themselves.
“Choice” field types create their own set of challenges. Choices may contain too few values, too many values, or ambiguous values that are unclear or subjective. For example, a lead source value labeled “Existing Customer” can mean different things in different contexts. Does it mean the prospect is the employee of an existing customer, or that an existing customer referred the prospect? Small ambiguities like these can create downstream process and reporting challenges. In these instances, some additional design conversations can unlock your data by creating crystal clear, unambiguous values that users can use confidently.
Good Data Hygiene & Design
Good data hygiene requires keeping similar data alongside similar data. A common example is Lead Source tracking. Organizations often combine marketing channels and campaign/content types in a single field. Values such as “Paid Search,” “Website,” and “Event” often exist in the same dropdown, even though they represent different data points. An event attendee may have arrived at the event landing page because of a paid search ad. In that case, they were both attracted by the Event (Lead Source) and arrived at the page through Paid Search (Traffic Source). Also, a value like Website alone is often too vague to be usable. When multiple dimensions are folded into a single field, automation, predictive analytics, and AI become less usable.
Good data design extends beyond values themselves. Field types matter just as much. Think of the Yes/No field type. For a value like “Completed Onboarding?” the answer is either yes or no. However, there are other applications where there is a third distinct answer: “We don’t know.” Think of calendar invites where “No Response” (blank) is quite different from “Declined” (No) or “Accepted” (Yes). With a Yes/No field (also called Boolean), there is no distinction between a “No” and “We don’t know.” For a critical field, that nuance can make a big difference on usability.
Special Attention for Sensitive Information
From a governance perspective, sensitive information requires special attention. Organizations handling patient information, financial data, Social Security numbers, or credit card information should put security controls in place so that only authorized users can access sensitive fields. Marketing consent is similarly important. Businesses should keep clear records of when and how consent was provided, including timestamps, opt-in/opt-out status, form details, and specific preferences that are collected each time consent records are updated.
Importance of Clean, Structured Data
As Copilot, AI, and other machine learning capabilities become more prevalent, the importance of clean, structured data continues to grow. AI can only use the data it receives. When data is well organized, clearly defined, and consistently captured, organizations can gain more value from AI, automation, reporting initiatives, and improved user experience and adoption.
Successful data design starts by cultivating an architect’s mindset, with deep consideration of how the data will be used now and in the future. This also requires ongoing monitoring and improvements. Periodic monitoring of processes, fields, mappings, and option sets can reveal quick wins. Others reveal deeper issues that nonetheless need to be addressed. Organizations that invest in these fundamentals create a stronger foundation for user adoption, reporting accuracy, automation, and future AI success.
How Forvis Mazars Can Help
Business Technology Services at Forvis Mazars can assist organizations with enterprise resource planning (ERP), CRM, advanced technology, and managed service solutions. If you have any questions or need assistance, please contact us.