Salesforce's New Sales Agent Already Built 60% of a Customer's Pipeline

Salesforce expanded Agentforce on September 11 with seven “job-ready” agents built for specific functions instead of one general-purpose assistant: Casey for customer service, Piper for inbound lead generation, Marshall for supply chain, and — the one that matters most for sales orgs — Hunter, an outbound-sales agent currently in pilot with general availability targeted for November. Salesforce says the platform has delivered 7 billion “Agentic Work Units” across Agentforce and Slack over two years, 3.2 billion of them in Q2 alone.

The customer numbers are the part worth sitting with. Salesforce cites Hunter building 60% of a customer’s sales pipeline, a help agent resolving 50% of chat inquiries, and a 4x increase in conversation volume from Piper at Asana. Hunter is also the first agent built on a new “long-horizon runtime” that lets it pursue a goal over days or weeks using memory and durable execution, rather than resetting after each interaction — the architectural piece that turns a chatbot into something closer to a persistent SDR.

This isn’t Salesforce’s first quantified proof point for packaged sales agents. Retailer Batteries Plus used Agentforce to stand up roughly 10 purpose-built agents aimed at a base of 100,000 commercial prospects — a list no human sales team could realistically work end to end — using agents to research, qualify, and hand off warm opportunities while reps closed. That’s the same pattern Hunter is now packaging as a single named product instead of a custom build.

The throughline across both is composition, not automation for its own sake: narrow, purpose-scoped agents assigned to specific pipeline stages, with humans still closing. For sales leaders evaluating whether “AI SDR” claims are real or vaporware, Hunter’s pilot-stage 60%-of-pipeline number, still unverified at broader scale, is the figure to track through its November GA.