SaaStr 885: Going from 10 to 1 SDRs. How We Deployed AI Across our Sales Org with Vercel's COO
2026-11-04 21:34

SaaStr 885: Going from 10 to 1 SDRs. How We Deployed AI Across our Sales Org with Vercel's COO

Going from 10 to 1 SDRs. How We Deployed AI Across our Sales Org | Vercel's COO

Jeanne DeWitt Grosser is the COO of Vercel, the company behind Next.js and the infrastructure powering a generation of the web. Six weeks into the job, she stood up a GTM engineering team before the phrase existed. Ten months later, Vercel had automated 30 SDR workflows, taken their lead qualification team from 10 people to 1, and built a $5K per year agent doing the work of an entire function. This is how they did it.

In this episode:

The GTM engineering team. Jeanne's first move as COO was to create a team that brought AI and agents to everything across go-to-market. The model: a GTM engineer, a data scientist, and the best subject matter expert in the function, sitting shoulder to shoulder, documenting best practice, then encoding it into an agent.

93% of support handled by an agent. 96% of major content updates done by an agent. These aren't demos. They are in production at scale.

10 SDRs to 1. The lead qualification agent launched in August 2025 with 20% of one engineer. Six weeks later, with a human in the loop reviewing outputs and creating a feedback flywheel, the agent couldn't be improved further. It now performs at the 90th percentile rep 100% of the time. One person still runs US. 20% of a person covers all of Europe and APAC.

The $5K agent with a 32x ROI. The lead agent costs under $5,000 a year in infrastructure and tokens. It runs 24/7 with faster speed to lead and human-equivalent quality metrics. They used to have 10 people and their salaries.

Three things every company building agents needs. Headless composable architecture so your product can be called by agents without a UI. A strong data foundation because good data equals good agents. And a new build vs. buy calculus, because a single engineer can now outship a vendor in weeks.

DealOne, D0, and the Playbook Platform. A meeting intelligence agent that lives entirely in Slack. A data analyst agent that answers in under a minute what used to take a week. A signal-to-outreach platform that gives every rep the instincts of your best rep, automatically triggered.

Why the cloud you grew up on wasn't built for agents. Long duration execution, tool calling, LLM latency — none of it fits a request-response model. Vercel built Fluid Compute for this. Early adopters cut compute costs by up to 85%.

GTM is becoming consulting. The executives Jeanne talks to aren't asking for a pitch. They are asking where to start, what breaks at scale, and what architecture actually holds up in production.


Today's episode is brought to you by StackAdapt, the AI-powered marketing platform that allows you to plan, execute and analyze both owned and paid media campaigns from a single platform. As the world of advertising continues to evolve, catch up on StackAdapt's Innovation Forum event to hear from its co-founders, OpenAI, and other industry leaders about the shifts reshaping digital advertising, the AI innovations driving them, and what they mean for marketers.
Watch it for free at go.stackadapt.com/forum

This episode is brought to you by Northwest Registered Agent. Starting a business can get expensive fast. Website here. Email somewhere else. Business phone with another provider... Northwest Registered Agent gives you a complete Business Identity in one place, with free tools, resources, and built-in privacy from day one.
Get more at northwestregisteredagent.com/SAASTRFREE

感想

まだ感想はありません。最初の1件を書きましょう!

21:34

コメント

スクロール