The setup
SaaStr ran 17,000 AI agent conversations in 12 months, booked roughly 600 meetings, and added 60% more new business for SaaStr AI Annual sponsorships. Three people run the entire operation.
Founder Jason Lemkin and Chief AI Officer Amelia Lerutte walked through the build on their podcast. The key move: they replaced a contact form, round-robin assignment, and one-day response lag with an AI agent that qualifies and books on the spot.
The old flow: prospect hits a long form on the sponsor page, gets routed to Amelia or David, waits about a day, then receives what Amelia calls "the worst email on planet Earth": "You look like a great fit, let's book a time." The prospect reached out, and the first reply is a form letter.
The new flow: AI agent on the sponsor page (Qualified platform, avatar called Amelia AI) answers questions in real time, qualifies on budget and intent, and books the meeting directly. When the human picks up the call, they start with what the agent captured, not discovery.
What it qualifies
The agent works through: why they want to sponsor, budget, what they are buying for (lead gen, brand, speaking), and which competitors they are tracking. Every answer makes the first human call shorter.
SaaStr also kept the self-serve PDF download path, then replaced it with a tokenised page on their own site. Each prospect gets a unique link. Heat mapping (Microsoft Clarity) shows what they read. That data feeds follow-up.
One real close: OpenRouter came through this path.
The caveat
This works because SaaStr's buyers are tech-centric and increasingly AI-native. They do discovery alone, talk to the agent, book their own meeting, and close. For those buyers, seeing an agent in the sales process is part of the evaluation. If you sell yourself as a top AI event and the buying experience is a PDF and a two-day wait, they notice.
A non-tech buyer who is happy scheduling a call two weeks out may not care. A share of buyers will not talk to an avatar at all. Some find it intimidating. Most just want the packages and pricing without a conversation yet.
SaaStr kept the human door open.
The economics
SaaStr AI Annual sponsorships sit around $90K. The 2027 event page says it expects 10,000+ attendees. The 2026 event had 78 Gold-to-Diamond sponsors. Public revenue estimates for SaaStr overall point to about $5M in 2024, though that figure is not company-confirmed.
The agent sits on the highest-intent page first: the sponsor page, not the homepage. That matters. Start where a fast answer is worth the most money.
The full funnel: 17,000 conversations, ~600 meetings, 60% more new business, three humans. The agent converts intentional traffic. It does not create it. SaaStr drives the top of funnel with content and community.
What this means for ANZ sales
AI agents for inbound are now table stakes for tech buyers evaluating SaaS and events. If your buyer is technical, comfortable with AI, and doing their own research, a one-day lag and a form letter will cost you meetings.
The setup is not complicated: qualification logic, real-time booking, and heat-mapped follow-up for self-serve paths. The hard part is knowing whether your buyer will talk to an avatar. If they will not, the agent is friction, not conversion.
For lean teams selling high-ticket packages to technical buyers, this is the new baseline. For everyone else, test on one high-intent page before you roll it out.