Actionable Metrics: A Test, Rewrites and Decision Rules

Most dashboards have more numbers than decisions. This guide gives a five-question test for actionable metrics, rewrites common vanity metrics into ones that lead somewhere, and shows the decision-rule format that turns a number on a screen into a change in what the team does.

What makes a metric actionable

An actionable metric is one where a change in the number tells you what to do next. A vanity metric looks good in a report and changes nothing: total sign-ups since launch, all-time page views, followers. The term comes from the lean startup movement, and the idea is simple: if nobody would make a different decision when the number moves, the number is decoration.

Vanity metrics are not false. They are usually cumulative, unsegmented or disconnected from a decision. The same raw data can almost always be turned into an actionable metric by changing how it is counted.

A five-question test

Ask these questions of every number on your dashboard:

1. Can it go down? Cumulative totals only rise. A metric that cannot fall cannot warn you.
2. Is it a rate or a ratio? "1,200 sign-ups" depends on traffic; "2.4% of visitors signed up" can be compared across weeks and channels.
3. Can you split it? A metric you can break down by channel, page, device or cohort tells you where to act.
4. Does it have an owner? Someone has to be responsible for moving it, or it will only be looked at.
5. Is there a decision waiting on it? You should be able to say "if it falls below X, we will do Y".

A metric that passes all five is actionable. One that fails the first two is almost always a vanity metric.

Audit a dashboard

The example audits a typical marketing dashboard. Of seven numbers, three are vanity metrics: total sign-ups since launch, all-time page views and social followers. They are not useless (they can show scale to investors or motivate a team), but they should not be in the place where weekly decisions are made. The four actionable ones are rates, can be split, and each has a clear response if it falls.

Rewrite vanity metrics

The table rewrites common vanity metrics into actionable ones. The pattern is always the same: replace a cumulative count with a rate over a period, add the denominator, and segment it. "Page views" becomes "pages per session on docs, by entry source"; "sign-ups" becomes "sign-up conversion rate by channel, weekly".

Same raw data, counted as a segmented rate over a period.

Write decision rules

The step most teams skip is writing down what the number will trigger. A decision rule has four parts:

If [metric] is [below / above threshold] for [duration], [owner] will [action].

For example: "If sign-up conversion from paid search is below 1.5% for two consecutive weeks, the growth lead pauses the lowest-converting ad group and reviews the landing page." Rules like this stop the weekly meeting from becoming a tour of numbers. They also expose metrics without a rule: if you cannot write one, the metric is probably vanity in your context.

Pair each rule with a leading indicator where possible, so the rule fires before the lagging outcome is lost.

Segment before you conclude

An average across all visitors often hides the action. A flat conversion rate can be a rising mobile rate and a falling desktop rate. Before acting on any metric, split it by the two or three dimensions most likely to differ (channel, device, new vs returning) or by a segment you care about. Segmentation is often what turns a metric from interesting into actionable, because it tells you where to look.

Actionable metrics in TapCub

TapCub reports are built around rates and filters. Every report can be filtered by channel, page, device, country and other dimensions, and the filters travel with the link, so a shared view shows exactly the segment you looked at (filters and sharing). A metric you define, such as a ratio of two events, can be saved and reused in charts and dashboards.

Decision rules can be backed by monitors: a monitor checks a built-in measure or a saved metric hourly, daily, weekly or monthly against a threshold, a period-over-period change, a forecast band or an anomaly rule, and notifies you when it fires (monitors). The analytics metrics guide and the dashboard guide cover which metrics to put where.

FAQ

What are actionable metrics?

Metrics where a change tells you what to do: usually rates or ratios over a period that can be segmented, have an owner and are tied to a decision.

What are vanity metrics?

Numbers that look impressive but do not guide decisions, typically cumulative totals such as all-time sign-ups, total page views or followers.

Are page views a vanity metric?

Total page views usually are. Pages per session on a given section, split by source, can be actionable if a decision depends on it.

How do I turn a vanity metric into an actionable one?

Replace the cumulative count with a rate over a period, add the denominator, segment it, and write a decision rule for it.

Should I remove vanity metrics from dashboards?

Move them out of the dashboard used for weekly decisions. They can stay in a separate overview for scale and morale, clearly labelled.