agentleFS
CharleneTStrainCharleneTStrain/product-ops-skills/skills/kpi-health-check/SKILL.md

kpi-health-check

Skill0 starsChanged 20 days agoMITView on GitHub

Compare the latest product KPIs against their baseline, target, and recent trend, flag what's genuinely off, and say what to check next — without over-reading normal noise. Use when someone pastes a dashboard export, weekly or monthly numbers, or a metrics table and asks "how are we doing," "anything weird this week," for a KPI review, or a metrics summary for a meeting.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use the agreed definitions (see the product-metrics-definitions skill).

    line 25
  2. Distinctive:

    Never state a cause that the data doesn't show.

    line 48
  3. Distinctive:

    Keep it to one screen. If everything is on track, the report is two lines long, and that's a good report.

    line 67

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

KPI health check

The point of a health check is to tell people where to look, not to explain every wiggle. Most weeks, most numbers are fine. Say so.

Before you start

If .agents/product-ops-context.md exists (or .claude/product-ops-context.md), read it first. Use the team, owners, tools, meetings, and metric definitions it already has, and only ask for what's missing or specific to this task. If it doesn't exist, go ahead with what you're given, and suggest running product-ops-context once so the next skill doesn't have to ask.

Step 1: Line up the numbers

For each KPI, put these side by side. Compute changes with actual arithmetic; never estimate.

KPIThis periodLast periodChangeSame period last year (if given)TargetRecent range

"Recent range" is the low and high over the last several periods, if the data allows. It shows what normal variation looks like.

Use the agreed definitions (see the product-metrics-definitions skill). If a number's definition is unclear, say so before judging it.

Step 2: Judge each KPI

Mark each one:

  • On track: within target and within its normal range
  • Watch: outside its normal range but not yet past target, or moving the wrong way for two or more periods in a row
  • Off track: past target in the wrong direction, or a sharp break from its normal range
  • Can't tell: missing data, a definition change, or too few periods to know

Be careful with:

  • Small numbers: a jump from 4 to 8 is +100% and may mean nothing
  • Known one-offs: holidays, outages, launches, pricing changes, tracking changes
  • Mix shifts: a total can move because the customer mix changed, not because behavior did

Step 3: Say what to check

For every Watch or Off track item, give one or two specific things to check, stated as questions, not conclusions:

  • "Did the drop come from one segment or all of them?"
  • "Did tracking change around the release on the 14th?"

Never state a cause that the data doesn't show. A likely explanation is a question to check.

Output

# KPI health check — [Period]

**Overall:** [one sentence, e.g. "Mostly steady; activation is the one to watch."]

## Needs attention
| KPI | Now | Change | Status | Check next |

## On track
- [KPI]: [value] ([change])

## Can't tell
- [KPI]: [why]

Keep it to one screen. If everything is on track, the report is two lines long, and that's a good report.

Related skills

  • product-metrics-definitions — agree what each number means first
  • data-request-router — when a flag needs a deeper data pull
  • quarterly-planning-prep — carry results into planning

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.