RunTheAgent
Playbook

Lead Research Agent

Build prospect lists of 25-50 companies matching your ICP with key contacts, company intel, and relevance scoring.

What You Will Get

Building a prospect list the manual way means hours of clicking through LinkedIn, company websites, Crunchbase, and industry directories. You end up with a spreadsheet of names that might fit your ICP, but half of them are outdated, irrelevant, or missing critical details like the right contact person or current company size.

The Lead Research Agent flips this process. You describe your ideal customer profile to your agent, and it uses tavily-search and agent-browser to scour the web for companies that match. For each company it finds, it pulls the key contact, their role, company size, recent funding, tech stack signals, and anything else that helps you qualify them. Then it scores each prospect on how well they match your ICP so you can prioritize the best-fit leads.

You end up with a structured list of 25-50 companies, each with enough context to write a personalized outreach message without doing any additional research. Run this weekly from WhatsApp, Telegram, or Slack to keep your pipeline full without spending your selling hours on research.

Setup Steps

Get your Lead Research Agent running in about 10 minutes.

1

Define Your Ideal Customer Profile

Message your agent and describe your ICP in plain language. Include industry, company size, geography, tech stack, funding stage, or any other criteria that matter. Be specific: 'B2B SaaS companies with 50-200 employees in the US that use Salesforce' is much better than 'tech companies.'

2

Enable Research Skills

Make sure tavily-search and agent-browser are active. These allow your agent to search across the web and visit company pages, job boards, and directories to gather intel that goes beyond basic firmographic data.

3

Set Your Output Format

Tell your agent what fields you need for each prospect. Common fields include company name, website, key contact name and title, email pattern, employee count, recent news, relevance score, and a one-line reason for the score. Your agent will structure every list this way going forward.

4

Choose Your Sources

Let your agent know which types of sources to prioritize. Job boards can signal buying intent, funding databases reveal growth-stage companies, and industry directories surface niche players. Your agent will weight its search accordingly.

5

Run Your First List

Ask your agent to build a list of 25 companies matching your ICP. It will search, visit pages, extract data, score each company, and deliver a structured list. Review the results and give feedback on which prospects were strong fits and which missed the mark.

6

Refine the Scoring

Based on your feedback, tell your agent to adjust the scoring criteria. Maybe you want to weigh recent funding higher, or deprioritize companies that already use a competitor. Each iteration makes the next list more accurate.

Tips and Best Practices

Start with 25, Scale to 50

Your first list should be smaller so you can review quality carefully. Once you are happy with the accuracy, increase to 50 companies per batch. Quality feedback early saves you from wading through mediocre leads later.

Feed Lists Directly into Cold Outreach

Once you have a list, forward it to your Cold Outreach Drafter agent. The research context carries over so your outreach emails will reference the same intel your Lead Research Agent found.

Run Weekly for Fresh Leads

Companies change fast. New funding rounds, leadership changes, and product launches can turn a cold lead into a hot one. Run a fresh list every week and your agent will surface companies that newly match your criteria.

Use Negative Filters

Tell your agent which types of companies to exclude. Existing customers, known competitors, companies under a certain size, or industries you do not serve. This keeps your lists clean and actionable from the start.

Frequently Asked Questions

Related Pages

Ready to get started?

Deploy your own OpenClaw instance in under 60 seconds. No VPS, no Docker, no SSH. Just your personal AI assistant, ready to work.

Starting at $24.50/mo. Everything included. 3-day money-back guarantee.

RunTheAgent
AParagonVenture

© 2026 RunTheAgent. All rights reserved.