SaaStr, a B2B SaaS company, recently reduced its number of AI sales agents from nearly 30 to about 20, resulting in a roughly fourfold increase in output, the company revealed this month. The decision came after the team realized they could no longer effectively manage additional AI agents that required human interaction, despite the agents running a real eight-figure business with live customers and invoices, according to saastr.com.
About a year ago, SaaStr deployed four distinct AI sales agents, each with unique roles and personalities: Agentforce focused on reviving ghosted leads using Salesforce data; Artisan handled warm outbound leads; Monaco targeted cold ideal customer profiles; and Qualified managed inbound conversions. However, as the company scaled, the outbound agents began to overlap in capabilities, making it feasible to consolidate their functions into fewer agents without losing effectiveness, saastr.com reported.
This consolidation reflects a broader trend in AI agent deployment where managing human-facing AI tools demands significant attention, context, and maintenance. By reducing the number of agents, SaaStr improved operational efficiency and output, demonstrating that fewer, more versatile AI agents can outperform a larger, more fragmented set. The move also highlights the challenges companies face in balancing AI automation with human oversight in sales processes.
SaaStr currently operates with three human employees supported by over 20 AI agents managing various sales functions. The company’s experience underscores the importance of optimizing AI agent deployment to maintain scalability and efficiency in B2B SaaS sales operations, as detailed in the latest update on saastr.com.