.png)
Meta-metrics answer one simple question: is analytics working as a system, or are we just producing output? They help you detect the usual failure modes early. Dashboards exist but are not used. Metrics exist but are not trusted. Changes take weeks, so the business routes around the analytics team.
A good meta-metrics set is intentionally small. If it takes more than a few minutes to explain or measure, it will not be used. The goal is to create a lightweight scoreboard that influences how you run analytics week by week.
Adoption tells you whether the steering set shows up where decisions are made. Track whether the right leaders use the agreed steering metrics in recurring meetings, and whether the key decision forums actually run on the steering set. If it isn’t embedded in a review cadence, it won’t stick.
Trust tells you whether people believe the numbers. Track how often the organization has “metric disputes” and how many critical metrics have a clear owner, definition, and validation path. Watch for workarounds: Excel exports, “my own numbers”, shadow logic.
Consistency tells you whether the organization is converging on one language. A practical indicator is the percentage of steering metrics that have exactly one agreed definition. This is where a metric store / published definitions pay off.
Speed tells you whether analytics can keep up with the business. Track time-to-answer for new steering questions and time-to-change for metric definitions. If changes take weeks, the business routes around analytics.
Impact tells you whether analytics improves outcomes, not output. Track value directly against your lighthouse wins. Tie at least a subset of wins to a measurable before/after.
Write down your meta-metrics as one short list. Keep it stable for at least one quarter.
| Category | Meta-metric (name) | Definition (what exactly is counted?) | How measured (source / method) | Cadence (weekly / monthly / quarterly) | Owner | Target (next quarter) | Current | Notes / actions if off-track |
|---|---|---|---|---|---|---|---|---|
| Adoption | ||||||||
| Trust | ||||||||
| Consistency | ||||||||
| Speed | ||||||||
| Impact |
If you only measure output (dashboards shipped, tickets closed), you will optimize for busy work. Meta-metrics keep the system honest and make it obvious where you need to invest next.