Behind the scenes of building Vercel’s first GTM Agent

Hey there,

This week I’m sharing the full breakdown of how we built Vercel’s first GTM agent. The results: 10 inbound SDRs → 1, $2M+ in savings, and a 32x ROI.

I joined Alex Lieberman on the Human in the Loop podcast to walk through the whole thing. The blog post is a cleaned-up version of that conversation with everything you need to build something similar.

Building Vercel’s First GTM Agent

The short version: we turned lead qualification from a slow slog into something that happens in minutes. The human’s job went from “sift through spam, do research, qualify, write email, send” to “review AI’s work, press send.”

A few other highlights:

Stated vs Revealed Preference — This is the number one trick I use when deciding where to spend GTM Engineering time. When someone says “we should do X,” I ask: how are you solving that today? If the answer is a wild, manual, painful process, you’ve found pay dirt.

Lessons from Building AI Agents for Financial Services — Nicolas Bustamante shared an article on building Fintool. My initial reaction: most of this applies cross-domain.

His big insight: “Context is the product.” Financial data comes in every format imaginable (SEC filings, earnings transcripts, press releases, market data) and the real work is normalizing all of it into clean, structured context the model can reason over.

That’s exactly what we’re doing with our internal knowledge base. CRM data, enrichment providers, Gong calls, support tickets, product usage. Different schemas, different update frequencies, different quality levels. The agent needs one thing: clean context (probably in markdown format).

Worth the full read if you’re building agents in any industry.

Shipping AI-Powered GTM workflows with Vercel — Thursday I'm chatting with Jeanne DeWitt Grosser (Vercel's COO) and Andrew Qu (creator of d0, our internal data agent). We're talking about more of the behind the scenes of how Vercel builds agents and the prototype to production pipeline.

That’s all for this week.

— Drew

Don’t miss the next one.