Back to Blog
AI & Automation

Clay + AI: Building Outbound Sequences That Actually Reply

March 1, 2025
8 min read
Clay + AI: Building Outbound Sequences That Actually Reply

A step-by-step walkthrough of the Clay workflow we use to research, enrich, and personalize 1,000 outbound sequences in under an hour.

The Problem With Old-School Outbound

Generic sequences. Low reply rates. Reps spending 3 hours per day on research that could be automated.

We fixed this with Clay.

The Clay Workflow

Step 1: Build Your ICP List

Source leads from Apollo, LinkedIn Sales Nav, or a custom scrape. Import into Clay.

Step 2: Enrich With Context

Use Clay's enrichment waterfall:

  • Company size, funding stage, tech stack
  • Recent LinkedIn posts from the prospect
  • Job change signals
  • Company news from the last 30 days

Step 3: AI Personalization

Prompt Claude or GPT-4 with enrichment data:

You are a cold email writer. Using the following context about [prospect], write a 2-sentence personalized first line that references something specific about their role or company.

Context: [enrichment data]

Step 4: Sequence in Instantly

Export to Instantly with:

  • Personalized first lines
  • Offer-specific body copy
  • Simple CTA ("15 min?", not "Would you be interested in scheduling a call at your earliest convenience?")

Benchmarks We Hit

  • 38% open rate
  • 22% reply rate
  • 8% meeting booked rate

Cold email isn't dead. Generic cold email is dead.

Found this useful? Let's talk about your GTM.

Book a Free Call →