the gtm engineer Building an AI agent-first GTM machine with Frank Sondors, CEO & Founder of Forge
Frank Sondors went from managing a fifty person sales team to building software that enables companies to scale their best salespeople. As CEO of Forge, he's built multiple interconnected products that cover the entire cold outbound stack, from email infrastructure and deliverability to lead generation and AI powered execution. Frank scaled Forge to $3 million ARR in under 12 months while hiring just three salespeople and using AI agents that book 500 meetings per month. In this conversation, Frank shares his contrarian approach to scaling revenue without scaling headcount, and we explore how AI agents are fundamentally changing the economics and strategies of outbound sales.
In this podcast, we discuss:
* The core problem with existing sales automation software tools and why Frank built Forge
* How the modern tech stack can enable companies to scale revenue without scaling headcount
* How Forge's AI agents achieve 2% reply rates on cold email
* Which sorts of products and use cases work with AI agents, and which do not
* How to to identify and automate repetitive work across your org
* A number of Frank’s favorite growth hacks that have helped his team scale to $3M ARR
* The other decisions and drivers behind Forge’s $0 to $3M ARR growth
Episode highlights:
* Frank explains that Forge built N8N workflows to automatically WhatsApp message new signups within two minutes, achieving 10x higher reply rates than email.
* Forge uses voice AI rather than SDRs to call leads who don't convert within 14 days in order to gather feedback on why they didn't purchase.
* To scale without hiring, Frank's team religiously automates repetitive tasks through a Slack channel called "N8N Ideas", where anyone can request automation. They prioritize automation ideas based on revenue impact and ship new workflows weekly.
* Frank's team runs campaigns where half the leads get AI written emails and half get human written emails to see which performs better. Forge then uses these results to inform future campaign copy.
* Frank suggests scraping and targeting the LinkedIn followers of your competitors because they have already shown interest in your category. Campaigns targeting these prospects consistently deliver better results than traditional intent signals like funding events or new hires, which everyone else is already targeting.
* Despite building outbound software, 80% of Forge's revenue comes from inbound. Frank attributes this to building in public by sharing daily updates about product and company progress, working US hours despite being based in Europe, and offering a website widget that lets prospects instantly call their sales team if they’re online.
Where to find Frank:
* LinkedIn: https://www.linkedin.com/in/franksondors/
* Forge
Transcript details:
(00:00) Intro
(02:41) Frank's background and the Forge origin story
(04:23) The early set of tactics that Forge automated and building an automation culture
(11:47) Overview of Forge's product ecosystem
(13:50) Using AI agents to book demos
(18:05) Forge’s agent success rates and where agents are less effective
(21:01) The role of real SDRs at an AI first company
(27:10) Automating WhatsApp follow ups
(28:47) How Frank thinks about agent prompting, context building, intent signals, and human-in-the-loop
(35:05) How Frank would think about finding winning messaging against competitors
(39:20) The best agents’ edge is their data layer
(47:08) Scaling to $3 million ARR in under 12 months and $1 million ARR without making a hire
(49:55) The impact of building in public
(52:25) Frank’s advice on building out go-to-market side of an org
(56:23) Favorite underrated software tool and conclusion
For inquiries about sponsoring the podcast and to recommend any guests, email noah@thegtmengineer.ai
This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit thegtmengineer.substack.com