vendor-analysis

name	vendor-analysis
description	Deep-dive analysis of a single vendor — financial health (10-Q / 10-K / D&B, ratings, recent earnings), operational performance (delivery, quality, NPS), technical fit (capability gap, roadmap alignment), risk (cyber, geographic, regulatory, ESG), and opportunity (consolidation, expansion, renegotiation). Outputs HTML report + Excel + executive recommendation. Use when the user says "vendor analysis", "supplier deep dive", "tell me about [vendor]", "review vendor X", "is X a strategic partner", or before a major renewal / consolidation / risk-review decision.
argument-hint	<vendor-name> [--depth quick|standard|deep] [--lens financial|operational|risk|all]

/vallor:vendor-analysis

Purpose

Before a renewal, a consolidation decision, an audit, or an executive briefing, the question is the same: who is this vendor, really? This skill produces a 360-degree view.

Load context

1. Read ~/.claude/plugins/config/vallor/CLAUDE.md:
   • ## Vendor / supplier classification — what tier is this vendor today?
   • ## Risk management → Supply risk dimensions — what we always track
   • ## Categories under management — which categories does this vendor serve?
2. Read the matter / vendor folder for prior reviews, contract briefs, negotiation history.
3. Apply the /vallor:design skill.

Ask the user for context (optional but encouraged)

Use the AskUserQuestion tool to gather optional context that materially changes the analysis. A vendor deep-dive for a renewal is different from one for a consolidation decision.

Up front (before research): Ask 1-3 short questions when not obvious from the inputs and profile:

• Decision driving this analysis: renewal / consolidation / risk review / audit response / new-vendor evaluation
• Depth (if not via --depth): quick (one-pager) / standard / deep (full report + appendices)
• Lens (if not via --lens): financial / operational / risk / strategic fit / all
• Specific concerns the user already has (e.g., recent missed SLA, leadership change, M&A rumor)
• Time horizon: 12-month decision / 3-year strategic / immediate (today)

Mid-flight (after the profile and initial research): Once you've pulled financials and operational data, surface concrete asks before guessing on:

• Conflicting signals (e.g., strong financials, weak delivery performance) — emphasize which
• Whether to model alternative scenarios (continue / consolidate / split / exit)
• Whether to include a recommendation or stay neutral (some users want the data, not the call)
• Cybersecurity / compliance depth — surface-level certifications or pull into a separate sub-analysis

Rules:

• Every question must offer a reasonable default the user can pick to skip ahead.
• Ask follow-ups when answers reveal more depth or a non-default choice.
• Cap a single round at ~3 questions.
• Skip when the user has signaled a fast / default run.

Workflow

Step 1 — Profile

Build the headline profile:

• Legal entity, HQ, geographic footprint
• Public vs private, parent company
• Annual revenue, headcount
• Categories supplied to us
• Annual spend with us (from spend-analysis if available)
• Our share of their book (rough estimate from public data)

Step 2 — Financial health

Pull and synthesize:

• Last 4 quarters of earnings (if public)
• D&B Paydex, credit rating, recent rating actions
• Recent acquisitions, divestitures, leadership changes
• Stock price / analyst sentiment (if public)
• Recent news (last 12 months) — material events

Tag financial pressure points and growth areas that affect our relationship.

Step 3 — Operational performance

For each category they supply:

• Delivery on-time-in-full (OTIF) trend
• Quality (PPM, returns, audit findings)
• Service incident history (if services / SaaS)
• SLA compliance (if applicable)
• Responsiveness (escalation cycle time)

Pull from CLM, ERP, ticketing systems if connected; otherwise ask the user for the data file.

Step 4 — Technical / capability fit

• Their stated capabilities vs what we use
• Roadmap alignment (their published roadmap vs our planned needs)
• Capability gaps (areas where they're not the best — and an alternative is growing)
• Innovation pipeline (R&D spend, patents, new products)

Step 5 — Risk

For each of our standard risk dimensions:

Dimension	Signal	Score (1-5)	Note
Financial	[D&B + earnings trend]
Geographic / geopolitical	[HQ + sourcing footprint]
Quality	[PPM trend, recalls]
Cybersecurity	[SOC 2 / ISO 27001 / breaches]
ESG / regulatory	[scope 1-3, EUDR, conflict-minerals, etc.]
Concentration (our share of their book)	[...]
Concentration (their share of our spend)	[...]

Step 6 — Opportunity

Three types:

• Consolidate — they could absorb work we have spread across other vendors.
• Renegotiate — they're in a weak position; we have leverage now.
• Diversify — they're a single-source risk; we should develop a backup.

Each opportunity scored: estimated dollar impact × effort × time.

Step 7 — Recommendation

A single executive recommendation:

• Keep + grow (strategic partner)
• Maintain (preferred supplier; nothing material to change)
• Renegotiate (open the contract; specific levers)
• Diversify (develop secondary source; reduce exposure)
• Exit (transition off; timeline)

Output

Save to: ~/.claude/plugins/config/vallor/vendor-analyses/<YYYY-MM-DD>-<vendor-slug>/

Files:

• report.html — interactive multi-section HTML (Overview · Financial · Operational · Risk · Opportunity · Recommendation). Chart.js for charts.
• data.xlsx — the underlying tables, sortable, finance-friendly.
• recommendation.md — one-page executive memo.

HTML report structure

Single page, six anchored sections (not tabs — this is a read-top-to-bottom doc):

1. Profile — masthead + headline cards (annual spend, our share of their book, tenure, current tier).
2. Financial health — earnings line chart, key metrics, recent events timeline.
3. Operational performance — OTIF / quality trend lines, SLA compliance bars.
4. Technical fit — capability radar (us-required vs vendor-delivered).
5. Risk — heat map across dimensions.
6. Recommendation — bordered callout with the recommendation and a single "What I'd do today" line.

Sources

Every external claim is sourced. Numbered list at the end. URL + access date. Tag confidence ([verified] / [reported] / [inferred]).

If a research connector (Westlaw, CourtListener, etc.) isn't available, web search is used and tagged. If neither is available, the report explicitly says so in the reviewer note — no silent supplementation.

Decision tree after delivery

1. Open the renegotiation — /vallor:negotiation-report builds the prep packet.
2. Run consolidation — /vallor:vendor-consolidation looks for overlap across the vendor base.
3. Update the procurement profile — promote / demote the tier if the analysis warrants.
4. Build the savings tracker — log the opportunity into the savings pipeline.
5. Stop here — analysis is saved.

Examples

/vallor:vendor-analysis Acme

/vallor:vendor-analysis Megacorp --depth deep --lens risk