Data Governance, Variables, and Data Quality
What this domain tests
Keeping data trustworthy and compliant. Concepts: variable classification (traffic vs. conversion vs. events), allocation and expiration settings for eVars, processing rules, VISTA (concept only), classifications and SAINT importers, data retention, user permissions, and data quality checks (missing values, case/whitespace inconsistencies, bot traffic, duplicate hits). Know the difference between classification (metadata enrichment) and processing rules (conditional modification at collection).
Traps to watch for
- Assuming eVars always persist; expiration and allocation must be set intentionally.
- Confusing classifications (add descriptive attributes to a value) with segments (subset of data).
- Overlooking case sensitivity and trailing spaces that fragment values.
- Believing processing rules can create new variables; they modify values conditionally.
Worked example
Marketing sends campaign codes inconsistently: email_q3, Email_Q3, email_q3 (trailing space). Fix: apply a processing rule to lowercase and trim, then use a classification to map codes to campaign names and channels. Result: one clean dimension value with rich metadata, no report rework.
Rule of thumb
Garbage in, permanent garbage out - analytics data is hard to retroactively fix. Lock naming conventions, apply processing rules at collection, and document expiration/allocation per variable.