Skip to content
GTM Agent Weekly

Turn a Sales Navigator saved search into scored, personalized Clay rows

This setup takes a targeted prospect list from LinkedIn, then verifies emails, scores fit and writes a personalized first line for each person. Reps start outreach from a ready list instead of a raw search.

Paste this into ChatGPT, Claude or your agent tool. See the prompt

  • Prospecting / Outbound
  • intermediate
  • SDR / BDR
  • Founder
Category
Prospecting / Outbound
Difficulty
Intermediate
Est. cost
Not stated by source

Problem

Sales Navigator builds precise ICP lists but does not enrich, verify emails, score or personalize. Doing that by hand eats most of an SDR's week.

Setup

An agency builds an ICP saved search and lead list in LinkedIn Sales Navigator, moves the list into a Clay table through a scraping tool or Clay's LinkedIn enrichment, then runs waterfall email verification, ICP scoring and AI-written first lines in Clay. Only verified, well-scored rows are formatted for import into the sending tool.

Prompt

Starter prompt by The GTM Agent Weekly, written from the setup described on this page. It is not the source's own prompt; see the source link below for their write-up.

Starter prompt · 23 lines
# ROLE
You are the scoring and personalization step for prospect rows in a table built from a
LinkedIn Sales Navigator search.

# INPUTS PER ROW
- {{PERSON}}: name, title, seniority, profile summary, recent activity if present.
- {{COMPANY}}: name, industry, headcount, description.
- {{EMAIL_STATUS}}: verified, risky or not found (from email verification).
- {{ICP}}: our ideal customer and buyer roles.

# TASK
1. Score ICP fit 1 to 5 with a one-line reason.
2. If fit is {{MIN_SCORE}} or higher and email is verified, write one first line: specific
   to the person or company, tied to a problem we solve, under {{MAX_WORDS}} words.
3. Otherwise mark the row "do not send" with the reason.

# OUTPUT FORMAT (JSON)
{"score": 0, "reason": "", "send": true, "first_line": ""}

# GUARDRAILS
- Only rows with verified emails and a good score go to the sending tool.
- No fake familiarity ("loved your post") unless the input shows it.
- Never invent achievements, metrics or news.

Replace every {{PLACEHOLDER}} before running. Works with any capable LLM and runner.

Tools

Steps

  1. Build a saved search with lead filters (titles, headcount, industry) and account filters (growth, technologies, funding).
  2. Keep each list to a few hundred tightly matched contacts and name it with the criteria and date.
  3. Move the list into Clay by scraper export, LinkedIn URLs or Clay's find-contacts-at-company.
  4. Run waterfall email verification, ICP scoring and AI first-line personalization.
  5. Export only verified, high-fit rows to the sending tool.

Results

Agency describes its production workflow and reports volumes and find rates; self-reported.

What goes wrong

Gotchas

  • Sales Navigator has no native export, so the list depends on a third-party tool.
  • List quality drops as you widen beyond the ICP core.
  • Scraping LinkedIn can violate its terms and risk the account (editor note).

Source and credit

formanorden.com ›

Forman Orden · vendor

Source dated 2026-10-11 · last reviewed 2026-10-11. Summarized in our own words; see the source for details.

Related.

Plan a rep's field day: scored stops, an optimized route and leave-behinds

For reps who visit accounts in person, this agent picks the best accounts nearby, plans an efficient route, and preps a brief and leave-behind page per stop. Afterward it logs visits and next steps in the CRM.

Works with any of SalesforceHubSpotGoogle Maps +5

  • Prospecting / Outbound
  • intermediate
  • Reference setup

Deliver a one-page account brief to the rep before every meeting

Before every sales meeting, this agent sends the rep a one-page brief on the account: recent news, past conversations and open questions. Reps walk in prepared without digging through notes.

Works with any of Google CalendarOutlookSalesforce +7

  • Prospecting / Outbound
  • intermediate
  • Reference setup