Claude + FullEnrich Tested: Full Account Map for Linear in 30 Seconds

@socialwithaayan ran an A/B test on Linear (a company with no stake in the outcome) using the same prompt ("build an account map for Linear and enrich contacts, work emails only")—one window with pure Claude, another with the FullEnrich connector attached. The verdict: the data source is what makes or breaks a sales intelligence workflow—pure Claude found only 6 executives, while with FullEnrich it produced a complete account map in 30 seconds.

Confirmed

• Setup takes just three steps: add a custom connector in Claude Connectors, paste the MCP URL, and log in to authorize (leave the OAuth field blank).
• Control group (pure Claude): found only 6 Linear leadership members, estimated headcount at 90-120, got zero verified work emails, and explicitly refused to guess based on email naming patterns.
• Test group (FullEnrich): pulled all 20 people at manager level and above (3 founders, COO, 7 heads, 1 director, 8 managers), found work emails for all 20, consuming 20 credits (1 credit per email).
• The account map tagged each contact with their buying role (decision maker / functional owner / champion), start date, and LinkedIn link, and auto-interpreted the org structure: no VP layer, a new EMEA team set up in 2026, and more.
• Connector limits: Claude only reads the first 10 rows directly; the remaining emails had to be manually attached via CSV export to complete the map. Claude also proactively warned against guessing remaining contacts from naming patterns.

Why it matters

• The test offers a quantifiable head-to-head showing the capability gap for LLMs with versus without external data sources: no matter how strong a model's reasoning, it can't produce usable sales intelligence without reliable data.
• Compared to the traditional process the author mocked (manually hunting decision-makers on LinkedIn, searching emails one by one, praying they don't bounce—"this isn't research, it's a second job"), the connector workflow compresses account mapping to the 30-second scale.
• The post also honestly documents the connector's limitations (only reading the first 10 rows directly), making it a useful reference for evaluating the practicality of such workflows.

2026-10-28 ~ 2026-10-28 · 7 related posts

• The old account-mapping grind: LinkedIn scraping and email hunting 'is a second job' — socialwithaayan · 2026-10-28
• [source] Test setup: same prompt on Linear, Claude alone vs. Claude with FullEnrich — socialwithaayan · 2026-10-28
• [source] Control group: Claude alone found 6 leaders, guessed 90-120 headcount, zero verified emails — socialwithaayan · 2026-10-28
• FullEnrich results: 20 manager-level contacts at Linear, 20 work emails, 20 credits — socialwithaayan · 2026-10-28
• Claude's account map output: buying roles, start dates and org signals in one pass — socialwithaayan · 2026-10-28
• Connector caveat: Claude reads only first 10 rows, rest must come via attached CSV — socialwithaayan · 2026-10-28
• [source] Claude + FullEnrich MCP: 30-second setup maps Linear's entire org with 20 verified emails — socialwithaayan · 2026-10-28