SaaStr runs sales on 20 AI agents, two Salesforce acquisitions
## From Shelfware to AI Hub in 10 Months Ten months ago, SaaStr's sales team had shrunk to three people. Two refused to log into Salesforce, even when paid $200 a month to do it. When they added Momentum (auto-logs sales calls to Salesforce), one rep quit over the activity tracking. Now SaaStr runs 20+ AI agents, all routing through Salesforce. The setup: Momentum and Attention for call intelligence, four AI SDR instances (Artisan running three campaigns, Monaco handling true outbound), Qualified for inbound, and Salesforce's own Agentforce managing win-backs. ## The Stack in Practice Momentum transcribes every call, extracts next steps, objections, competitor mentions, and pushes structured data to Salesforce. No rep touches it. Artisan sent 15,000 messages in 100 days across ticket sales, sponsorship outreach, and VIP reactivation. Response rates: 5-7%. Monaco booked meetings with top AI companies from day one. Qualified powers the inbound AI agent on saastr.com. Full Salesforce context on every visitor: past attendance, engagement history, ICP fit. Not a chatbot asking the same qualification questions. It knows who it's talking to. Agentforce handles the mess: 1,000 warm inbound sponsor leads that got routed to reps and received zero follow-up. Ever. High-intent leads ghosted by the team. ## Salesforce Buying the Best Agents Pattern worth noting: Salesforce just acquired Momentum. Salesforce just acquired Qualified. They are pulling the best sales agents in the stack in-house. If you are building on Salesforce or evaluating AI SDR tools, that acquisition strategy matters. ## What This Means for Sales Teams The AI agent hub problem is real once you pass five agents. Where does the data go? How do agents talk to each other? When they conflict, who wins? SaaStr's answer: Salesforce became the system of record because it was the only one capable of handling it. Not by design. By necessity. For sales leaders evaluating AI tooling: the CRM question is not about features anymore. It is about whether your CRM can be the central nervous system for autonomous agents. Most cannot. Salesforce can, and they are acquiring the best agents to prove it. Worth asking: what does your stack look like when you have 4 AI SDRs, auto-logged calls, inbound qualification, and win-back automation all running at once? Who is the source of truth? How do you avoid agents stepping on each other? SaaStr went from paying for shelfware to running their entire GTM motion through it. The difference: AI agents that need somewhere to live.