Lemkin says agents closed $2.7M in leads reps never touched
Three humans, 20+ agents in daily production, and revenue that he says moved from -19% year over year to +47%.
Jason Lemkin published the breakdown behind his claim that SaaStr doubled revenue with AI agents. He says the company runs on three humans and 20+ AI agents in daily production, and that revenue went from -19% year over year to +47%.
The headline number comes from the least glamorous place in the business. Lemkin says the sales agent, built on Agentforce, worked around 1,000 warm leads that were sitting untouched in Salesforce, got a 72% open rate on outreach, closed $2.7M in deals that reps never touched and nurtured $3.5M in pipeline.
The cheap agents found the leaks
Two other agents get names in his write-up. 10K, described as SaaStr's AI VP of Marketing, runs daily revenue forecasting, the newsletter to a roughly 450K database, and LinkedIn and X ads end to end. Lemkin says 10K recommended a 15% ticket price cut that drove about 40% attendance growth, ran a Marketo to Salesforce Marketing Cloud migration for about $14 in compute, and found two customers still being billed $300 a month for a product SaaStr killed six years ago. Nobody there knew.
do not put AI on your A leads. Put it on the B leads your humans will never follow up with. That is where the real revenue is.
vibe-coded apps need daily maintenance. Agents drift, data changes, models update, integrations flake. We QA ours every day.
QB, the AI VP of Customer Success, manages 150+ customers end to end, and Lemkin says she sent 83 personalized sponsor emails at 12.23am one night. Combined monthly cost to run 10K and QB, he says, is $257. A separate inbound agent on Qualified booked 614 meetings and handled 402,000 chats for SaaStr Annual alone. In a follow-up post he says the agents did 17,000 customer conversations and booked 600 meetings in the past 12 months.
The part that is a daily job
Lemkin is blunt that none of this is set and forget. He says agents drift, data changes, models update and integrations flake, so his team QAs theirs every day.
These are one founder's numbers, self-reported, not audited. But the shape of the claim is the useful bit. The revenue did not come from replacing people doing good work, it came from the B-tier follow-up and the billing audit nobody had hours for. And anything touching live pipeline comes with a maintenance bill paid in attention, not dollars.