Data quality assessment and management plans
Key Takeaway
After you assess how data is used and where its quality falls short, the first step to improving it is to develop a data management plan. Salesforce features then put that plan into action.
Data quality is the degree to which data meets its users' expectations for its intended use, so data must be "fit for business purpose" to be reliable. Bad data is linked to lost revenue, missing or inaccurate insights, wasted time, poor customer service and decreased adoption by reps. To improve data quality, you first assess your records against dimensions such as age and completeness. Next, a data management plan sets the standards for your data. Salesforce features such as required fields, validation rules and Duplicate Management then put the plan into action.
How to assess data quality
- Learn how the company uses customer data to support its business objectives. Ask teams four questions:
- What are your business objectives?
- What customer data do those objectives need?
- How do you use that data?
- Where is that data stored?
- Measure the data in Salesforce against key dimensions: age, completeness, accuracy, consistency, duplication and usage.
Many departments often use the same customer data, but each has its own process or source of truth, such as spreadsheets. The result is duplicates, incomplete records and stale data.
Each dimension has its own way to measure it.
| Dimension | How to measure it |
|---|---|
| Age | Run a report on the "Last Modified Date" of records to see what percentage were updated recently. |
| Completeness | List the fields each business use needs. Then run a report that shows the percentage of blanks in those fields. A data quality app from AgentExchange can also help. |
| Accuracy | Install a data quality app from AgentExchange that matches your records against a trusted source. |
| Consistency | Run a report on the values in date, currency, state, country, region and language fields. Count how many variations exist for a single value. |
| Duplication | Use the Duplicate Management features in Salesforce and a duplicate detection app from AgentExchange. |
| Usage | Review whether the data is being used in reports, dashboards and apps. |
You often can't prove that a value is correct, but rules or algorithms can detect when a value might be incorrect. For example, an email address like none@none.com that shows up on a very large share of records points to invalid contact data.
You don't have to build every assessment report yourself. You can install a prebuilt package from AgentExchange, such as the Data Quality Analysis Dashboards App. The app adds ready-made dashboards, such as High Level Data Quality Analysis and Account, Contact & Opportunity Data Quality. For example, these dashboards can show opportunities that are missing key fields like lead source and next steps.
Lena is the new admin at a home insurance agency. She asks the sales, service and marketing teams the four questions. She learns that marketing keeps its own spreadsheet of webinar sign-ups. Back in Salesforce, a report on Last Modified Date shows that just 30% of contacts were updated in the past year, so Age is a problem. A report on the Country field finds the United States entered as "United States", "USA" and "U.S." One value has three variations, so Consistency is a problem too.