| name | clay-core-workflow-b |
| description | Use Claygent AI research and AI-powered personalization to generate outreach copy from enriched data.
Use when writing personalized email openers, running Claygent research prompts,
or configuring AI columns for campaign personalization at scale.
Trigger with phrases like "clay AI personalization", "claygent research",
"clay outreach copy", "clay secondary workflow", "clay AI column".
|
| allowed-tools | Read, Write, Edit, Bash(curl:*), Bash(npm:*), Grep |
| version | 1.14.0 |
| license | MIT |
| author | Jeremy Longshore <jeremy@intentsolutions.io> |
| tags | ["saas","clay","workflow","ai","claygent"] |
| compatibility | Designed for Claude Code, also compatible with Codex and OpenClaw |
Clay Core Workflow B: Claygent AI Research & Personalization
Overview
Complements the enrichment pipeline (clay-core-workflow-a) with AI-powered research and personalization. Uses Claygent (Clay's built-in AI research agent powered by GPT-4) to scrape websites, extract insights, and generate personalized outreach copy for each prospect. 30% of Clay customers use Claygent daily, generating 500K+ research tasks per day.
Prerequisites
- Completed
clay-core-workflow-a with enriched table
- Clay Pro plan or higher (Claygent requires Pro+)
- Understanding of prompt engineering basics
Instructions
Step 1: Add a Claygent Research Column
In your Clay table with enriched leads:
- Click + Add Column > Use AI (Claygent)
- Choose model: Claygent Neon (best for data extraction and formatting)
- Write your research prompt referencing table columns:
Research {{Company Name}} ({{domain}}) and find:
1. Their most recent funding round (amount, date, investors)
2. Any recent product launches or major announcements from the last 6 months
3. Their primary competitors
Return results as structured data. If information is not found, return "Not found" for that field.
- Enable Auto-run on new rows
Step 2: Configure Multi-Output Claygent (Neon Model)
Claygent Neon can extract multiple data points into separate columns from a single run:
Research the company at {{domain}} and extract:
Output 1 (Recent News): The most notable company news from the last 90 days. One sentence.
Output 2 (Tech Stack): List the main technologies they use (check job postings, BuiltWith, Wappalyzer data).
Output 3 (Pain Points): Based on their Glassdoor reviews and recent job postings, identify likely operational pain points.
Output 4 (Competitor): Name their primary competitor.
Map each output to a separate column for downstream use in personalization.
Step 3: Build a Personalized Email Opener Column
Add an AI column (not Claygent -- use the faster AI model for text generation):
You are a sales copywriter. Write a personalized 2-sentence email opener for {{first_name}} at {{Company Name}}.
Context about the prospect:
- Title: {{Job Title}}
- Company size: {{Employee Count}} employees
- Industry: {{Industry}}
- Recent news: {{Recent News}}
- Tech stack: {{Tech Stack}}
Rules:
- Reference one specific fact about their company (not generic)
- Do NOT use "I noticed" or "I came across" (overused)
- Keep it under 40 words
- Sound human, not AI-generated
- End with a natural transition to your value prop
