Skip to main content

Smart agents

L
Written by Levon Adamyan
Updated this week

Smart Agents are AI-powered custom fields that extract specific information from LinkedIn profiles using natural language queries you define.

Plan requirement: Smart agents are available on Starter and higher plans (not Free).

How they differ from smart columns

While smart columns are pre-built analysis types (ICP scoring, email validation, etc.), smart agents let you define any question in plain English and have AI analyze each lead's LinkedIn profile to answer it.

How to create a smart agent

  1. Name your column - Give it a descriptive name (e.g., "Budget Authority").

  2. Define your query - Write what you want to extract in natural language. Example: "Does this person likely have budget authority for software purchases based on their title and role?"

  3. Select leads - Choose which leads to process.

  4. Run - The AI fetches LinkedIn data, analyzes it, and populates results.

Accuracy

  • 90-95% for factual queries (current job title, company name)

  • 80-90% for inference-based queries (budget authority, decision-making power)

  • 70-85% for complex analysis (leadership style, cultural fit)

Processing speed and cost

  • Speed: 3-8 seconds per lead on average

  • Cost: 1-5 credits per lead depending on query complexity

  • Failed processing costs zero credits

Example use cases

  • Sales: "Does this person manage a team?" / "What technologies does their company likely use?"

  • Recruiting: "How many years of management experience does this person have?"

  • Research: "What industry certifications does this person hold?"

Limitations

Smart agents cannot access private information, predict future behavior, fetch real-time data beyond LinkedIn profiles, or bypass LinkedIn's premium restrictions.

Did this answer your question?