Build a prospect list from an ICP

When to use

• The user has no list yet and wants one built from criteria.
• "Find me 50 heads of marketing at Series A SaaS companies in Canada."
• If the user already has target companies, use find-decision-makers instead.

Before you start

1. Check the connection: call get_account with {} or run npx ravn whoami. Note lookups left this window.
2. search_people needs your own provider key connected — run ravn keys add pdl first; these tools run on your own provider account. If it returns 403 provider_key_required, stop and tell the human to add their key. search_business_email_by_name and email verification are ours and need no key. Every call is one lookup; misses are free.
3. Agree on a target size with the user.

Steps

1. Turn the ICP into filters. Lists mean OR within a field and AND across fields. - title: the job titles, with common variants. - seniority: Intern, Entry, Senior, Manager, Director, VP, Head, C Suite, Owner, Partner. - industry, country, state, city, department. - employee_range: 1-10, 11-50, 51-200, 201-500, 501-1000, 1001-5000, 5001+. - Exclusions: exclude_company_domain (existing customers), exclude_industry, exclude_seniority.
2. Run a small test page first to check the filters return the right people.

{
  "title": ["VP Operations", "Head of Operations", "Director of Operations"],
  "industry": "logistics",
  "state": "Texas",
  "employee_range": ["51-200"],
  "exclude_company_domain": ["currentcustomer.com"],
  "limit": 5
}

1. Show the user the five results and total. If they are off target, adjust the filters before spending more. If status is not_found, loosen one filter at a time.
2. Page through with limit up to 25 and the cursor set to the previous next_cursor, until you reach the target size or next_cursor is empty.
3. Each profile may include business_email and business_email_risk_score. For people without an email, call search_business_email_by_name with their name and company.domain:

{ "first_name": "Maria", "last_name": "Lopez", "company": "lonestarfreight.com" }

1. Verify every email you plan to use with one validate_emails call (up to 1,000 addresses):

{ "emails": ["maria.lopez@lonestarfreight.com", "j.chen@haulpoint.com"] }

1. Keep valid. Mark risky and unknown for review and leave them out of the send list.
2. Save the list as a CSV.

Output

A CSV with one row per person:

• full_name, title, seniority, company_name, company_domain, company_size, city, state, country, linkedin_url
• email, email_source (search_people or search_business_email_by_name)
• rw_verdict (valid, invalid, risky, unknown) and rw_risk_score
• rw_status (ok or not_found) for the email step

Also report: filters used, total matches available, people pulled, verified emails, and lookups used.

Rules

• People who opted out are removed from the data already. Also remove anyone the user knows has asked not to be contacted. See /privacy#optout.
• Drop invalid and risky emails before the list is used for outreach.
• No FCRA uses: never build lists to screen people for credit, employment, insurance or housing.
• This is for building a working prospect list, not bulk exporting the database. Pull what the campaign needs.
• On a 429 fair_use_exhausted or a 403 provider_key_required, stop and tell the human.

Example prompt

Build me a list of 40 operations leaders (director and up) at 51-200 person logistics companies in Texas, skip our current customers in customers.txt, with verified emails.