Stats
Actions
Tags
From deal-diligence
build, audit, and maintain company-specific kpi trees that decompose a budget, forecast, or operating target into causal drivers and atomic inputs. use when a user wants to break revenue, margin, opex, headcount, cash flow, or other plan lines into measurable drivers; create a diligence-ready budget framework; translate a model into operating inputs; or define what should be tracked weekly and monthly post-close. especially useful in late diligence, annual planning, monthly business reviews, and post-acquisition operating cadence design.
How this skill is triggered — by the user, by Claude, or both
Slash command
/deal-diligence:kpi-tree-builderThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Build KPI trees that connect financial outcomes to measurable operational inputs.
Build KPI trees that connect financial outcomes to measurable operational inputs.
The purpose of this skill is to turn a budget, forecast, or target metric into a usable operating architecture. The output must help the user:
Always decompose:
Outcome -> driver -> sub-driver -> atomic input
Stop only when the input is:
Do not stop at vague labels like "sales productivity", "retention", or "utilization". Keep decomposing until each branch can be defined operationally.
This skill owns:
This skill is responsible for the structure of the tree itself. Keep the work focused on decomposition, classification, reconciliation, and tracking architecture.
If the user provides a model or budget tree:
Preserve useful structure, but do not mirror a management model blindly if the structure is not causal.
Use this when the user is evaluating whether a management budget is credible.
In this mode:
Always ask:
Use this when the user wants a tree that can be tracked over time.
In this mode:
Always ask:
Follow this sequence.
Start with the outcome the user cares about:
State:
Choose the right decomposition based on the business model.
Common patterns are in references/decomposition-patterns.md.
If the business is hybrid, build separate trees by revenue or cost stream first, then roll them up.
For B2B SaaS, recurring software, or other GTM-driven businesses, combine the economic decomposition in references/decomposition-patterns.md with the GTM metric system in references/example-output-gtm-kpi-tree.md.
For these businesses, do not stop the tree at generic branches such as:
Instead, decompose those branches into explicit GTM metrics when relevant, including:
ARR funnel metrics:
Funnel and pipeline metrics:
Retention and health metrics:
Efficiency and economics metrics:
Team productivity metrics:
Use these metrics only where they improve causal clarity. Do not force every metric into every tree. Prefer the smallest tree that still explains the outcome and supports management action.
Every major branch must be classified into one of:
This is mandatory. It makes later tracking and diagnostic work consistent.
For each branch, keep drilling down until inputs are:
Use references/node-classification.md to decide whether the decomposition has gone deep enough.
For GTM-driven businesses, a node is not deep enough if it cannot be mapped to a defined operational metric or formula.
Examples of valid GTM atomic inputs:
Avoid stopping at labels like:
Those labels are too broad unless they are decomposed into measurable sub-metrics.
For every node, assign:
The tree must reconcile bottom-up and top-down.
Always:
Translate the tree into a tracking cadence.
Typical rule:
Use references/tracking-architecture.md to specify cadence, owners, and management actions.
Default to this structure.
State:
Present as an indented tree.
For formatting, labeling, and depth, use references/example-output-revenue-tree.md as the default example.
When the company is a B2B SaaS or GTM-driven business, also use references/example-output-gtm-kpi-tree.md as a default example for full-funnel and operating-metric decomposition.
For each node, include:
If in diligence mode, include:
Recommend a focused set of:
The weekly list must only contain early-warning indicators.
Do not preserve a hierarchy just because it exists in a deck or model. Rebuild around economic logic.
Distinguish:
Do not collapse booked, live, billable, and recognized value into one node.
Do not hide them inside net retention if the user is trying to manage the business.
If segment, product, geography, channel, or cohort mix matters, create a mix branch.
External effects such as FX, regulation, or one-time timing items should not be mixed into management-controlled performance.
A KPI tree without owners is incomplete.
If a node cannot be measured, say so and propose the closest practical proxy.
driver-tree decomposes an investment thesis into causal drivers for IC memo purposes — revenue, cost, capital, and competitive dynamics — and maps each driver to NTBs and MOIC outcomes. kpi-tree-builder operates downstream of that: it converts the same causal structure into a trackable operating architecture with owners, cadences, and management actions. In a PE deal workflow, run driver-tree during diligence to build the thesis structure, then run kpi-tree-builder post-close to convert it into the operating model management tracks.
ntb-diligence identifies what must be true for the investment thesis to hold and flags which assumptions are unconfirmed. The diligence mode of kpi-tree-builder is the mechanism for resolving those gaps — the atomic inputs in the KPI tree are the measurable things that confirm or refute each NTB. When ntb-diligence has produced a GAP item, the kpi-tree-builder diligence view should map that gap to the specific tree node that resolves it.
gtm-metrics-analyzer is the measurement and calculation companion to this skill.
Use kpi-tree-builder to design the operating architecture:
Use gtm-metrics-analyzer to calculate and interpret the GTM metrics that sit inside those nodes:
This skill may name the metric nodes that should exist in the tree, but it should not perform the full metric-calculation workflow owned by gtm-metrics-analyzer.
A good output should let the user answer:
npx claudepluginhub ian-lawrence423/claude-skills --plugin deal-diligenceProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Searches MemPalace before answering questions about past work, people, projects, or prior decisions. Returns verbatim stored content instead of guessing from model memory.