These cookies are required for the website to function properly. They ensure security, enable basic features like page navigation, and store user session data. You cannot disable these cookies.
- Social Media Analytics
- Features
- Metric governance
Metric governance
Define measures and inspect calculated results.
OverviewApproved Definitions×
Social Media Analytics · Metric governance
New
Approved Definitions
Metric definitions
| Name | Grain | Unit | Aggregation method | Version | Status | |
|---|---|---|---|---|---|---|
| Definition 301 | GRN-129 | UNT-786 | Sum | Rev A | Approved | |
| Definition 921 | GRN-332 | UNT-859 | Count distinct | Rev B | Approved | |
| Definition 405 | GRN-276 | UNT-191 | Ratio of sums | Rev C | Approved | |
| Definition 164 | GRN-818 | UNT-940 | Last as of | Rev D | Approved | |
| Definition 363 | GRN-523 | UNT-708 | Median | Rev E | Approved |
6 of 6 records
Definition 301Metric definitions · MD-87906
SumGrainGRN-129
UnitUNT-786
VersionRev A
StatusApproved
What you can do
Agree on the meaning of a metric
A metric definitions record holds the measure name, grain, numerator and denominator expressions, unit, aggregation method and version. Its status separates Draft, Approved and Retired definitions. The Analytics Lead should make that definition understandable before analysts compare values across profiles. Two readings called engagement can mean different things if their denominator or grain differs.
Metric results stores the resulting value and period with the definition, source profile and operation run. Its status distinguishes Calculated, Suppressed and Invalid results, with a reason where appropriate. The analyst can inspect the numerator, denominator and source row count instead of treating a visible value as self-explanatory.
How the work is organized
- Definition. Record the metric grain, expressions, unit, aggregation method and version.
- Observation. Review source readings for the same metric and reporting window.
- Result. Keep a persisted value with its definition, profile, run and reason.
What to agree during setup
Decide who reviews a definition before it becomes Approved. Use the version to make formula changes visible to people comparing historical results. Agree how to handle a zero denominator and when a result should be Suppressed or Invalid. The reason should allow a reader to distinguish a missing input from a deliberate exclusion.
Choose a window and dimension key convention. A daily observation, a monthly profile result and a cohort comparison can all use the same words while describing different populations. Make the intended grain explicit in the definition and in the report discussion. If a channel definition changes, use its effective dates when deciding which metric definition applies.
Check the handoff on an actual case
Create a draft engagement definition and have the Analytics Lead review its expressions, aggregation method and version. Choose a profile with accepted facts and an observation linked to a covering snapshot. Compare the observation window with a result window and inspect the result’s numerator, denominator, source row count, status and operation run. A second person should be able to explain why that value belongs to the approved definition.
Repeat the check with a missing denominator. The analyst should record a result status and reason that prevent the number from being read as a valid engagement rate. Keep the original reading and review context so the team can resolve the source problem without silently replacing the old explanation.
What your team sees
- Approved Definitions — Find the measure versions available for reviewed comparison.
- Metric Results Review — Inspect calculated, suppressed and invalid values with their reasons.
- Observation Windows — Check which dated source readings informed the discussion.
Follow the work through
Use metric analysis for result inspection and profile measurement to review for the wider lifecycle.
What it keeps track of
Data model- Metric definitions
- Approved metric definitions with grain, expressions and aggregation method.
- Metric observations
- Observation-level metric readings tied to a snapshot and source.
- Metric results
- Calculated metric results persisted once per approved definition and window.
- Operation runs
- Workflow execution runs with frozen revision, input hash and result identity.
- Social channels
- Configured measurement and experiment definitions with effective dating.
- Social profiles
- Measured web properties with lifecycle status and observed outcome fields.
Who uses it
Permissions- Analytics Lead
- , Data Analyst, Independent Reviewer
Set up Social Media Analytics your way.
Start free, then use Proto to change fields, steps and approvals before your team goes live.