HG Insights|Phoenix

Skill: HG Tool Chain — Account Research

The repeatable account-research playbook every top AE runs in their head — now Claude runs it the same way every time.

Overview

The repeatable account-research playbook your top AE uses, encoded for Claude. Start from firmographic (size + parent disambiguation), narrow with technographic (do they run anything in our category?), size the deal with spend, find the buyer with FAI, name the humans with contact search, time the play with contracts, and ground it with SEC filings — each step gating the next so you don't waste calls on accounts that won't qualify.

Use cases

  • Account research that's the same quality as your best AE on their best day

    The chain executes deterministically: size, fit, spend, buyer, contacts, timing, public filings. No 'I forgot to check intent', no 'I jumped straight to contacts and got generic IT VPs'. Every account brief touches the same seven decision points in the same order.

  • New AEs ramped on the methodology without a 6-month shadowing curve

    A new hire installs this skill into Claude and produces account research that follows the same gating logic senior AEs use — no 'figure out which tools to call when' learning curve. The methodology is the skill.

View full skill

HG Tool Chain — Account Research

When to use

  • Authoring a new account-research workflow.
  • Auditing an existing workflow that calls tools in an unprincipled order.

The canonical chain

company_firmographic
  → company_technographic
  → company_spend
  → company_fai
  → contact_search
  → company_contracts
  → sec_filing_section (public companies only)

Each arrow is a gate: do not run the next step until the previous step's output justifies it.

Why this order

  1. company_firmographic first. Every later step needs an hg_id and a parent/subsidiary decision. Skip this and you risk the #1138 31-char hg_id failure (see hg-firmographic).
  2. company_technographic next. Tells you the stack. Gates whether the workflow has anything to say at all — a company with no technographic footprint in your category isn't a target.
  3. company_spend third. Sizes the opportunity. Gates whether the deal is meaningful (a 50-employee company with $200K modeled IT spend isn't an enterprise play).
  4. company_fai before any contact lookup. Tells you which department to target. See hg-fai-vs-contact-search.
  5. contact_search with the FAI-informed title pattern. Yields named humans with the right org context.
  6. company_contracts — pull renewal-window context for the discovery call. Skip this if the deliverable doesn't need it.
  7. sec_filing_section for public companies. Skip if firmographic shows the company is private — sec_filing_section will return nothing useful and burns 1 credit.

Where to drop a step

  • No company_spend when the workflow is purely about technographic fit (e.g., a competitive-replacement play).
  • No company_contracts when the workflow is about discovery (the renewal window matters for displacement, not for awareness).
  • No sec_filing_section for private companies, or when the workflow doesn't need filings.
  • No contact_search for territory-sweep workflows that only need account-level summaries.

Where to add a step

  • company_intent can replace or precede company_technographic when the workflow's primary signal is "what are they researching".
  • web_search as a fallback after company_firmographic for non-US filers without SEC presence; see hg-fallback-to-web-search.

Common pitfalls

  1. Calling contact_search before FAI. Generic title-pattern matches with no department grounding.
  2. Calling sec_filing_section before firmographic. Wastes credits on private companies.
  3. Re-running company_firmographic mid-chain. Cache the result from step 1; don't re-fetch.
  4. Running the entire chain on a 500-account territory. Pre-filter by firmographic first; reserve the full chain for the top 10-20.

Reference