All skills
deanpeters avatar

/company-intel

@b68bf96

Research a company, industry, or competitor set using web search and seven analytical lenses. Use when you need structured intel that feeds downstream PM skills.

Use this Skill: https://skilld.dev/gh/deanpeters/product-manager-skills/company-intel

This session only. Nothing lands on disk.

examplesexecutive-signal-refresh-saas.md

≈1.3k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Company Intel — Executive Signal Refresh Example (SaaS)

All companies, executives, quotes, figures, and URLs are fictional. SaaS sibling of executive-signal-refresh.md, using the suite's SaaS surrogate universe: the target is Vantiga, the enterprise software suite whose "Vantiga FSM / Vantiga Go" mid-market push appears throughout the market-intelligence examples. Where the industrial refresh read patents and segment P&Ls, a SaaS refresh reads ARR language, net-retention framing, and product-org signals — same pattern, different tells.

Trigger: "Run company-intel on Vantiga again — it's been a quarter, and their FSM unit keeps showing up in our deals."


What's Changed Since August 2026

1. "Vantiga Go" vanished from the earnings narrative

  • Then: the August run recorded Go (the mid-market FSM tier, launched March) as a named growth initiative — two consecutive calls had cited "early Go momentum."
  • Now: the Q3 call contains zero mentions of Go; mid-market appears once, in an analyst's question the CFO answered with suite-attach language instead — "our motion is strongest where the Vantiga platform is already the system of record" (Q3 call, Nov 2026, transcript URL).
  • Reading: the mid-market experiment is being starved or sunset — the answer-dodge pattern (asked about Go, answered about suite attach) is the classic pre-sunset tell — Inference.
  • Confidence: medium-high (one quarter of silence + a dodge; a second silent quarter would make it high)

2. Net-retention framing quietly replaced new-logo framing

  • Then: August baseline logged their growth story as new-logo counts by segment.
  • Now: Q3 materials lead with net revenue retention and multi-product attach rates; new-logo counts moved to an appendix slide — Fact (deck structure comparison, URLs).
  • Reading: growth is shifting from acquisition to expansion — consistent with starving Go: you don't need a mid-market front door if the strategy is deepening existing enterprise accounts — Inference.
  • Confidence: high (structural change in their own materials, corroborates shift #1)

Dropped Language

  • "Land-and-expand for the mid-market" — present in three consecutive quarters of materials through August; absent from Q3 deck and call entirely. Paired with Go's disappearance, the mid-market thesis looks retired without an announcement.
  • "Self-serve" — four mentions in the August investor materials, zero in Q3. For a suite company, self-serve vocabulary disappearing is the GTM telling you the sales motion won.

Section 9 delta (Strategic Signals)

  • Hiring: FSM-unit PM postings down from 6 to 1; the remaining posting reports into the platform org, not the FSM unit — Fact (job boards, Nov 2026). PM job postings are culture documents: the reporting-line change says the unit lost its product autonomy — Inference.
  • Leadership: the GM hired in 2025 to run the mid-market push moved to a "strategic advisor" title in October — Fact (their announcement). The soft-landing title within two quarters of the narrative shift completes the pattern.
  • Pricing page: Go tier still listed and purchasable — Fact (page, Nov 2026). The storefront outliving the strategy is normal; expect packaging consolidation within two quarters — Assumption (pattern-based; the tracker's watchlist owns it).

Sections with no material change

Sections 2 (revenue model), 3 (enterprise customer base), 7 (competitors): no significant change since August 2026.

What This Means for Fieldlight (perspective note)

The August read treated Vantiga Go as a live mid-market threat with low evidence quality. This refresh resolves it: the threat is receding by their own choice. Deals where "Vantiga is already in the building" remain hard; greenfield mid-market deals against them should get easier — update the battle-card "They win when" line accordingly — Inference.


Why this example works

  • Dropped Language carried both headlines again — but the SaaS tells differ from the industrial ones: there it was a transformation program and the word "platform"; here it's a tier name, "land-and-expand," and "self-serve." The pattern (what leaders stop saying) is universal; the vocabulary to watch is domain-specific — which is exactly why this sibling exists.
  • A metrics-framing shift was read as strategy, not finance. New-logo → NRR framing is the kind of change finance readers shrug at and strategy readers act on; the refresh treats earnings materials as strategy documents, per the skill's core teaching.
  • The org signal decoded the product signal. A PM posting's reporting line — not its existence — revealed the FSM unit's lost autonomy. Product-org heuristics doing forensic work.
  • The refresh closed a loop the suite opened: the snapshot's "Vantiga: evidence quality low, revisit" assumption gets its resolution here, and the output names the exact battle-card line to update — intelligence ending, as always, by naming the artifact it changes.

Source: SKILL.md on GitHub

1 warning1mo3 checks · Risk SAFE
  • Gen Agent Trust Hub1mo

    This skill acts as a research engine for company and market intelligence, gathering data via web searches. It is generally safe for use, although it has a low risk of indirect prompt injection because it processes untrusted content from the internet that could contain instructions meant to trick the AI.

  • Socket1mo

    No alerts

  • Snyk1mo

    Risk: MEDIUM · 1 issue

Signed by skilld at b68bf96. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Activeupdated 2 months ago
argument-hint
[company, industry, or competitor set]
type
workflow
theme
market-intelligence
Other metadata
intent
Act as a research engine that builds deep, structured understanding of companies, industries, and competitor sets through **seven analytical lenses** — financial landscape, market offer, product portfolio, competitive dynamics, rising trends, PM implications, and strategic signals (patents, hiring, leadership). Produces a stable, structured output that downstream skills consume to generate battlecards, SWOT analyses, positioning statements, PESTEL assessments, and market sizing. Supports four entry points: single company, industry/sector, named competitor set, or company + "discover my competitors" (researches the company first, identifies likely competitors, confirms the list, then runs the full competitor set analysis).
best_for
[
  "Building deep company or industry knowledge before a client engagement or workshop",
  "Generating structured research that feeds battlecards, SWOT, positioning, or PESTEL",
  "Running a competitive scan across 3-5 companies with cross-company comparison",
  "Refreshing intel quarterly on companies you track"
]
scenarios
[
  "Run company-intel on [company or their domain, e.g. helix-motion.com]",
  "Give me intel on the clinical data management industry",
  "Compare [Company A], [Company B], and [Company C]",
  "Run company-intel on [company] competitors",
  "Refresh my intel on [company] — what's changed since last quarter?"
]
estimated_time
30-60 min

README badge

README badge for deanpeters/product-manager-skills/company-intel