kpi-tree
Skill6 starsChanged 4 months agoMITView on GitHub
Defines the north-star metric, breaks it into 3-5 drivers with owners and leading indicators, specs the one-page weekly dashboard, and names the single metric that matters right now. Use when reporting is a pile of numbers nobody acts on, or when the team cannot agree what winning looks like this quarter.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- Ingramradical235/anty-frameworktheir own version, first seen Mar 20, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- Ingramradical235/anty-framework★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository29 other files it gives its agents.
- ad-scriptsSkill
- battlecardSkill
- brand-voiceSkill
- campaign-debriefSkill
- case-study-writerSkill
- channel-pickerSkill
- cmoSkill
- competitor-teardownSkill
- content-calendarSkill
- creative-briefSkill
- customer-deep-diveSkill
- email-sequenceSkill
- experiment-backlogSkill
- funnel-mapSkill
- hook-factorySkill
- landing-pageSkill
- launch-planSkill
- lead-magnet-builderSkill
- marketing-auditSkill
- market-research-sprintSkill
- message-mapSkill
- objection-mapSkill
- offer-builderSkill
- onboardSkill
- positioning-workshopSkill
- pricing-testSkill
- repurposeSkill
- research-to-briefSkill
- weekly-reviewSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 25
Update Numbers that matter in marketing-brief.md — north star, drivers with owners, baselines date-stamped, and the current one metric.
- Distinctive: line 32
Every number on the dashboard must be able to change a Monday decision.
- Distinctive: line 34
If the one metric is hard to name, the tree is too big.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.
Files
KPI Tree — the numbers that run the week
Most marketing dashboards are graveyards: forty metrics, no decisions. A KPI tree is the opposite — one north star that tracks value delivered, a handful of drivers that multiply into it, and a weekly page that says what to do differently on Monday. If a number cannot change a decision, it does not make the tree.
Inputs
- The business goal and the numbers currently tracked, with baselines where known: $ARGUMENTS
- From
marketing-brief.mdif present: Numbers that matter (current north star and baselines), the offer and price (unit economics hang off them), and the primary channel.
Do this
- Pick the north star: the one number that best tracks value delivered to customers AND moves revenue when marketing does its job. Test every candidate against gaming — if the team maxed this number and revenue stayed flat, it is the wrong star. Follower counts and impressions only qualify if the model genuinely monetises attention.
- Break the north star into 3–5 drivers, as arithmetic where possible — pipeline value = conversations × qualification rate × win rate × average deal, for instance. Arithmetic exposes which driver is actually the constraint.
- Attach each driver to an activity the team actually runs, and to an owner. A driver nobody owns is a hope.
- Mark each node leading or lagging, and pair every lagging number with the leading one that predicts it — revenue lags, conversations lead. Leading numbers run the week; lagging numbers judge the month.
- Spec the weekly one-page dashboard, with the source written down per number. Anything unmeasurable today goes on an instrumentation list, not the dashboard:
- North-star trend against target.
- Each driver with this week's number, last week's, and the delta.
- One experiment readout with its decision.
- Name the ONE metric that matters right now: the driver furthest from target with the strongest link to money. One, not three — this is the number the week gets organised around. Inside Claude Code, the metrics-architect agent runs the deep version of this pass.
- Update Numbers that matter in
marketing-brief.md— north star, drivers with owners, baselines date-stamped, and the current one metric.
Output
The KPI tree (north star, drivers with owners and their leading indicators), the weekly dashboard spec with sources, the instrumentation list, and the one metric that matters right now — saved into marketing-brief.md.
Rules
- Baselines come from real data or get marked UNKNOWN. A tree grown from invented baselines produces confident nonsense.
- Every number on the dashboard must be able to change a Monday decision. If it cannot, cut it.
- Date-stamp every baseline and target. Undated numbers rot into folklore.
- If the one metric is hard to name, the tree is too big. Prune until it is obvious.
- The tree serves the constraint. When the constraint moves, re-run this skill — do not let last quarter's tree run this quarter's week.
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
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.

