In today's episode, I chat with Joe Deville, founder of Clay Works of Art, about building creative B2B outbound campaigns for SaaS and IT startups before Series B.
We explore his unique data collection approach using government databases like sam.gov, HIPAA violation websites, and public contract data to find intent signals. Joe shares two brilliant campaigns: an automated Reddit agent that mentions Ignite on cybersecurity threads (generating 7 meetings), and targeting government contract winners who lack required security certifications. He reveals his controversial take that personalization doesn't work at scale, preferring message-market fit over individual customization, and uses tools like Octave and Twain to craft copy that resonates with 15-20 people instead of one. Joe discusses his journey from IT desktop support to Microsoft Azure admin to SDR at Jamf, why sales rejection was 10x harder than coding, and his prediction that everyone will have an agentic AI "Pikachu" using LangChain for knowledge management, data analysis, and campaign optimization.
Enjoy 🙂
(
00:00) Introduction to Outbound Wizards
(
00:26) What Clay Works of Art Does and Client Focus
(
01:32) The Process: Data Collection and Positioning
(
04:07) Creative Campaign 1: Automated Reddit Agent for Ignite
(
05:35) Creative Campaign 2: Government Contract Winners
(
06:55) Copywriting Philosophy: Why Personalization Doesn't Work
(
08:41) Journey from IT Admin to Sales (10x Harder Than Coding)
(
12:01) Future of GTM: Agentic AI as Your Right-Hand Man
(
13:42) When Deep Fakes Became Real (The New Source of Truth)
(
14:56) Closing and Contact Information
🔗 CONNECT WITH SAURAV
👥
LinkedIn
📧 Email - saurabh@salesrobot.co
🙏 LEAVE A REVIEW If you enjoyed listening to the podcast, we'd love for you to leave a 5-star review on Apple Podcasts to help others discover the show :)
👋🏼 GET IN TOUCH You can also reach out with any feedback, ideas or thoughts about the lessons you've learned from the episodes.
What is The GTM Engineer Podcast?
GTMEs didn't think they would evolve this way. They just kept automating things until it became their whole job. Now the tooling is moving faster than anyone can keep up with, there's no real playbook, and half the advice online is either too vague or written by someone who hasn't actually done it. This podcast is for you if you're in the middle of figuring it out - real workflows, numbers, and mistakes from expert GTMEs at startups, mid-market, and enterprise. If you're wiring up Clay, rethinking your outbound stack, or trying to make AI actually useful in your pipeline, welcome to the The GTM Engineer Podcast.