North Star Metric

Define a single, customer-centric North Star Metric and the 3–5 input metrics that drive it (a metrics constellation / input tree). The Evermuse twist: the North Star must reflect the value customers actually describe getting in real conversations — not a revenue proxy or a vanity number. You ground the value definition in quotes, then pick the metric that best captures it.

Step 0 — Relevance & availability

Confirm this is a metrics-definition task and Evermuse is connected. If the tools aren't present, produce a framework-only NSM labeled ⚠ ungrounded — Evermuse not connected and tell the user to authorize the MCP. See ${CLAUDE_PLUGIN_ROOT}/skills/using-evermuse/SKILL.md Step 0.

Evermuse Grounding (required)

Follow ${CLAUDE_PLUGIN_ROOT}/skills/using-evermuse/SKILL.md.

Required — one parallel batch of searches. Verify the product (Rule 1), then fire in a single parallel batch:

• 3–4 evidence searches, each worded from a different angle (limit up to 50). Evidence comes back rich and varied — expect large, useful result sets.
• one guidance search and one context search (limit up to 50). These are usually sparse or empty; run them anyway and note when they're thin.

Read each response's digest — it reports how many more results exist. Use judgment on whether a query is worth pulling deeper (raise limit toward the 100 max and/or page with next_offset to avoid repeats), weighing payload size, remaining context, task complexity, and the value of the data. For deep pulls, consider spawning sub-agents — instruct them to return every citation with the same metadata the tools return (url, who_said_it, meeting_name, created_at) so you can still cite.

Optional — considered use. Once grounded, reach for the other tools only when they add value: a quote-angled search for verbatim voice, read_source for a single deep dive, inline citations (references/citations.md), and add_source to save the deliverable. Sources, citations, and saving are optional — not required.

Instructions

You are a metrics strategist defining the North Star for $ARGUMENTS. Where the source method reaches for "business context you provide," use Evermuse evidence (how customers describe value) as the primary input.

NSM is NOT multiple metrics, a revenue/LTV metric, an OKR, or a strategy. NSM IS a single customer-centric KPI reflecting the value customers get and leading long-term business success.

Step 1 — Classify the business game

Determine which game the product plays; it shapes the North Star family:

• Attention — time customers spend (Spotify, YouTube).
• Transaction — exchanges between customer and platform (Uber, Airbnb, Amazon).
• Productivity — how efficiently someone completes work or reaches a goal (Notion, Canva, Loom). Ground the pick: which framing matches the value customers actually describe? Cite it.

Step 2 — Define the North Star from evidenced value

State the single metric that captures the core value customers named. Validate against the 7 criteria: (1) easy to understand, (2) customer-centric, (3) reflects sustainable value/habit, (4) aligned to vision, (5) quantitative, (6) actionable, (7) leading indicator. Show the metric passing each — and tie criterion (2) directly to a customer quote about value.

Step 3 — Build the input-metric tree (3–5)

Define the input metrics that most directly drive the North Star. Each input should be easier to move short-term, provably contribute to the North Star outcome, and point to where teams optimize. Draw the tree: North Star = f(input₁, input₂, …). For each input, prefer a good-metric shape — a ratio or rate, comparative over time, and behavior-changing ("if it won't change what you do, it's a bad metric"). Distinguish leading from lagging, and note where a qualitative signal (customer complaints, sentiment from the corpus) is the earliest leading indicator.

Step 4 — Guardrails (brief)

Name 1–2 health guardrails so optimizing the North Star doesn't quietly harm the customer (e.g. quality, latency, churn). This is the useful half of a metrics dashboard — not a full ops dashboard; for that, keep it out of scope.

Deliverable format

# North Star — [product]

**Business game:** [Attention / Transaction / Productivity] — because [evidenced value, [`1`](URL)].

**North Star Metric:** [metric]
**Why it captures value:** > "[customer quote about the value]" — [attribution] [`2`](URL)
**7-criteria check:** [✓ understandable · ✓ customer-centric [`2`](URL) · ✓ sustainable · ✓ vision-aligned [`3`](URL) · ✓ quantitative · ✓ actionable · ✓ leading]

**Input-metric tree**  (North Star = f(inputs))
| Input metric | Shape (ratio/rate) | How it drives NSM | Lead/Lag |
|---|---|---|---|
| [input 1] | [rate] | [link, cited [`4`](URL)] | Leading |

**Guardrails:** [health metric 1], [health metric 2].

Honesty when evidence is thin

If customers don't clearly articulate the value, say so — a North Star built on a guessed value definition is fragile. Flag it as a hypothesis and suggest interviews (/evermuse:interview if available) to confirm the value the metric is meant to track.

Save & hand off

After saving, offer: "Wire this into the GTM plan's metrics?" (/evermuse:gtm-strategy), or "Define the ICP whose value this measures?" (/evermuse:ideal-customer-profile).

Further reading

• Adapted from phuryn/pm-skills north-star-metric, absorbing the input-tree half of metrics-dashboard (MIT); grounding is Evermuse-native.
• The North Star Framework 101
• AARRR (Pirate) Metrics
• Are You Tracking the Right Metrics?