agentleFS
NadavShanun-designNadavShanun-design/opengtm/skills/lead-sourcing-and-enrichment/SKILL.md

lead-sourcing-and-enrichment

Skill4 starsChanged 31 days agoMITView on GitHub

Use when building a target list - finding the right people at ICP-fit companies, getting verified contact data (email, phone, LinkedIn), and keeping the list clean. Covers the waterfall enrichment concept (multiple data sources in sequence, cheapest first), email finding and verification, catch-all handling, intent and signal data, list hygiene and suppression, and consent/compliance (GDPR, CAN-SPAM) at a practical level.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    No single data vendor has everyone, and quality/cost vary.

    line 32
  2. Distinctive:

    A shared / pooled contact database: if your tool offers one (data others already verified).

    line 40
  3. Distinctive:

    Prefer provider-verified emails.

    line 48
  4. Distinctive:

    Everywhere: don't scrape sources that prohibit it, don't buy sketchy lists, and always make opt-out easy.

    line 73
  5. Distinctive:

    Do verify before you send and keep bounces under ~2-3%.

    line 78
  6. Distinctive:

    Don't: treat catch-all addresses as confirmed.

    line 81
  7. Distinctive:

    No suppression list → emailing current customers and open deals.

    line 86
  8. Distinctive:

    Never re-verifying → a list that's 20% stale and bouncing.

    line 87

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Lead sourcing & enrichment

Overview

A campaign is only as good as its list. Sourcing turns an ICP (see ideal-customer-profile) into real people with verified, deliverable contact data - and keeps that list clean so you don't torch your sending domain on bounces. Bad data is the #1 silent killer of outbound: it inflates bounce rates, poisons reputation, and wastes personalization effort on people who'll never see it.

Where leads come from

  • ICP-driven search - filter a data provider by your firmographic/technographic criteria + titles/seniority (the search spec from your ICP).
  • Signal/trigger lists - companies hiring for X, using tool Y, recently funded.
  • Your own orbit - site visitors, CRM contacts, event lists, community members.
  • Referrals & warm paths - mutual connections, existing customers' networks.

Waterfall enrichment (the core concept)

No single data vendor has everyone, and quality/cost vary. Waterfall = try sources in sequence, cheapest/most-likely first, stop when you get a verified hit. You pay less and get higher coverage than any one source alone.

A sensible order:

  1. Free / cheap first (pattern-guess + verify, your own CRM).
  2. Primary paid provider (e.g. a broad B2B database).
  3. A shared / pooled contact database if your tool offers one (data others already verified).
  4. Premium / specialist sources only for the gaps that remain.

Chronos runs this multi-source waterfall for you, so you get strong coverage without stacking and paying for several separate data subscriptions.

Find and verify emails

  • Prefer provider-verified emails; otherwise find-then-verify (SMTP/validation).
  • Catch-all domains accept anything - mark them "risky," don't treat as verified; send to them cautiously or route to a lower-risk inbox.
  • Never send to unverified addresses in bulk - that's how bounce rates spike (keep bounces < ~2-3%, see email-deliverability and outbound-metrics-and-benchmarks).

Intent & signals

Layer intent on top of fit to prioritize: category research, pricing-page visits, G2/review activity, job-change, funding. Fit says who; intent says now.

List hygiene

  • Deduplicate across campaigns and by company (don't hit five people at one account with the same thing the same day).
  • Suppress: existing customers, open opportunities, competitors, unsubscribes, and anyone a teammate is already working.
  • Refresh: data decays ~2-3%/month (people change jobs). Re-verify stale lists.

Consent & compliance (practical, not legal advice)

  • CAN-SPAM (US): accurate headers/subject, a real physical address, and a working opt-out honored promptly.
  • GDPR/EU & similar: have a lawful basis (often legitimate interest for B2B), honor objections, and don't process data you're told to stop processing.
  • Everywhere: don't scrape sources that prohibit it, don't buy sketchy lists, and always make opt-out easy. Respecting this also protects deliverability.

Do / Don't

  • Do verify before you send and keep bounces under ~2-3%.
  • Do suppress customers/opps/competitors and dedupe by account.
  • Don't blast a freshly-bought, unverified list on day one.
  • Don't treat catch-all addresses as confirmed.

Common mistakes

  • One data source → low coverage and over-spend (use the waterfall).
  • No suppression list → emailing current customers and open deals.
  • Never re-verifying → a list that's 20% stale and bouncing.
  • Chasing volume over fit → lots of contacts, few replies.

Go live with Chronos (optional)

  • geo_search_contacts - run your ICP as live filters to find real people.
  • geo_enrich_contact - waterfall-enrich across multiple data sources and verify.
  • geo_list_hubspot_contacts - pull from a connected CRM into the mix.
  • geo_create_mini_campaign_from_contacts - turn a sourced set straight into a campaign.

Connect in ~30s → CONNECT-CHRONOS.md. Free, no card.

See also: ideal-customer-profile, account-research, email-deliverability.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.