Choose Monitoring Methods and Cadence
Learning Objectives
After completing this unit, you’ll be able to:
- Describe different monitoring approaches and when each is appropriate.
- Determine how frequently to monitor data based on risk, change, and business impact.
- Explain how multiple monitoring methods work together to support stewardship and governance.
Monitoring Is Not One-Size-Fits-All
Different business processes and supporting data require different levels of monitoring. The right approach depends on how the data is used, how frequently it changes, and the risk associated with errors or drift.
Some monitoring activities are intentional and time-bound. For example, you should validate improvements after a cleanup effort. Others are ongoing, designed to detect gradual degradation or unexpected changes as systems, integrations, and user behaviors evolve.
Effective monitoring combines multiple methods because each answers a different question. Reports show the current state, profiling reveals trends over time, audit history explains what changed, automated alerts identify when action is needed, and stakeholder feedback validates the business impact.
Decide How Often to Monitor
Monitoring cadence should be driven by business context rather than technical convenience. Common patterns include:
-
Before-and-after monitoring to validate the impact of a manual cleanup or enrichment effort
-
Short-term automated monitoring of newly created or updated objects, fields, or validation rules to assess adoption after a process or configuration change
-
Periodic monitoring to detect data drift, emerging outliers, or changes in value distributions over time
Higher-risk data such as those that drive service level agreements (SLAs), routing, compliance, or automation typically warrants more frequent monitoring.
A Practical Monitoring Framework
The table below outlines common monitoring approaches and when they’re most effective.
Method |
What You Learn |
When to Use |
|---|---|---|
Stakeholder interviews and targeted reviews |
Where users struggle, lose trust, or work around the system. For example, NTO support agents exporting cases to spreadsheets because account data is incomplete or inconsistent |
Early discovery, post-launch validation, prioritization |
Reports and dashboards |
A point-in-time view of data quality and business activity. For example, NTO sees a high number of Contacts missing email addresses or Accounts without assigned owners. |
Regular reviews, leadership visibility |
Scheduled data profiling |
Preserved data and metadata statistics snapshots to detect trends, emerging outliers, and data drift over time. For example, Luna notices placeholder email addresses steadily increasing after a new integration or completeness declining month over month. |
After major changes; and for ongoing monitoring of data, metadata, and org health such as data quality compliance, system usage limits, completeness of metadata context, and security or operational risks |
Field history and audit logs |
Show who changed data or metadata, what changed, and when the change occurred. For example, Luna discovers that multiple automations are overwriting the same contact field after a flow deployment. |
Root cause analysis; adoption tracking |
Flow and automation |
When data crosses thresholds and needs action. For example, NTO triggers an alert when opportunity records are created without required fields or when duplicate emails exceed a set threshold |
High-risk fields and time-sensitive processes |
These approaches are complementary. Together, they provide both evidence and early warning signals.
How NTO Chooses the Right Monitoring Approach
At NTO, Luna chooses the right monitoring method based on the type of change and the level of risk.
After introducing new fields to support service workflows, she uses reports and dashboards to review newly created and updated records each day for a month. This helps confirm that users are adopting the fields and entering data correctly. When something looks off, she follows up with targeted reviews to understand where users are getting stuck.
For established, business-critical fields such as those supporting SLAs for business customers, Luna relies on automated data profiling to monitor trends and detect drift over time. She complements this with dashboards to track current performance and ensure the data continues to meet expectations as usage scales.
By adjusting both the monitoring method and cadence based on risk and impact, NTO avoids over-monitoring low-risk data while maintaining confidence where it matters most.
Unit Recap
Choosing the right monitoring methods and cadence helps organizations balance effort with value. By combining interviews, dashboards, profiling, audit history, and automation, teams can detect issues early, prioritize action, and support effective data stewardship.
In the next unit, learn how data quality KPIs and monitoring outputs guide action and long-term governance.