SaaStr, the B2B SaaS media and events platform founded by Jason Lemkin, is running what it calls an "8-figure" business heading toward $20 million revenue in 2026 with three humans and 21+ AI agents in production.
The agents handle inbound leads, meeting booking, lead reactivation, account research, CRM updates, invoicing, and collections. SaaStr reports agents generated $4.8 million in pipeline and $2.4 million in closed-won revenue, with one inbound agent booking 614 meetings at roughly $85,000 average deal value. These are self-reported figures.
The operating model is the story, but the market implication is bigger: agents are now evaluating and selecting vendors. Lemkin's rule is simple: if the AI VP Revenue wants to trial a tool, SaaStr trials it. If the agent says no, they pass.
This mirrors the developer-led buying shift Twilio rode in the 2010s, except the buyer is now an agent, not a human. For vendors, that means optimising for machine-readable positioning: authoritative third-party references, clean product data, integration documentation, security specs, pricing signals, and strong category association. The agent does not care about your brand story. It wants data it can parse and compare.
Competitive pressure is coming from AI-native sales platforms, CRM vendors adding autonomous agents (Salesforce has already shipped Agentforce SDR), and incumbent SaaS providers embedding agentic workflows. The available data does not confirm SaaStr's sales leadership team, headcount breakdown, or ANZ operations, but the company also runs a $200 million AI investment fund.
For ANZ sales teams, the takeaway is tactical: if your prospects are deploying agents for vendor research and shortlisting, your positioning needs to work for both humans and machines. That means structured data, not just compelling narratives. Agent-mediated procurement is not theoretical anymore. It is happening at scale, and it is shifting how software gets bought.