BB kpi-tree-design-coach
Coach a CEO, COO, CFO, Head of Strategy, Head of Operations, or VP-level functional leader through designing or rebuilding a KPI tree (also known as North Star Metric tree, OKR tree, value driver tree, metric tree, or key results tree) for a company, business unit, or function. KPI tree design is one of the highest-leverage strategy artifacts but most companies do it badly — they have a "metric soup" of 50 dashboards instead of a hierarchical decomposition of the one metric that matters into the operational levers people actually pull. Covers the foundational concepts (input metrics vs output metrics; leading vs lagging; lead-time of the metric; metric ownership vs metric awareness; the difference between a North Star Metric and a KPI tree), the North Star Metric selection (single-metric vs constellation; the criteria — measures customer value created, predicts long-term revenue, captures the business model), the tree structure (root → 1st-level drivers → 2nd-level drivers → operational metrics; multiplicative vs additive vs composite relationships; the 3-4 levels rule), the metric-tree archetypes by business model (B2B SaaS subscription, B2B SaaS PLG, B2C subscription, B2C transactional/e-commerce, marketplace, ad-supported media, fintech, healthcare/regulated), the operational principle (each leaf metric must be ownable by a specific person and movable on a sub-90-day timeframe), the cadence and ritual design (weekly business review, monthly operating review, quarterly strategy review — what each consumes from the tree), the integration with OKRs (the tree provides denominator metrics; OKRs are quarterly initiatives that move them; KRs are not metrics in themselves), the integration with planning (the tree is the spine of the operating plan and budget), the most-common failure modes (metric soup; vanity-metrics in the tree; too many leaves; metric ownership ambiguity; trees that ignore unit economics; trees that ignore quality; LTV/CAC as North Star is wrong; acti
As a process B 76/100 · Nearly there — weak spots: running it twice, progress reporting
How to improve
- Shorten the description to 1024 characters.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 2. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 2997 chars, limit 1024 - warning
body-longSKILL.md body ≈ 5327 tokens (recommended < 5000); move details to references/ - note
description-budgetdescription takes 2997 of the ~15000-char shared budget for all skills
Process rating: all ten parameters 76/100
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 9 mutating operations with no state check
- 60Result and completion. Output format stated, no completion criterion
- 70When it triggers. States when to use, but not when not to
- 70Inputs and preconditions. Inputs and preconditions are listed
- 70Failures and branches. 10 branches
- 70Execution cost. Instruction body is 5327 tokens
- 85Steps. 169 steps, 1 vague phrases
- 100Tools and files. No external tools needed
- 100Consistency. Name and required fields are in place
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 18 top-level sections: this looks like several domains in one skill
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 2997: 120–800 characters recommended
- +1No license
- +2Single-language instructions
- +5Description quotes 22 example trigger phrases
- +4Structure: 63 headings
- +3Step-by-step instructions: 169 items
- +3Output format is stated explicitly
- +4Has examples (1 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 51.