Salesforce introduced Agentforce Help Agent on March 6, a solution that combines the core components of customer service operations into a single environment. The new product addresses a common pain point for support teams: fragmented knowledge bases, business processes, and channel systems that undermine the business value of AI agents.
A unified platform that handles more than Q&A
Agentforce Help Agent is built on the company's Agentforce 360 Platform. Beyond answering questions, it performs follow-up tasks such as order management, appointment scheduling, and account management. Organizations can extend the platform to handle a wider range of customer requests without stitching together separate tools or workflows.
A pay-per-resolution pricing model
The most striking feature is the pricing structure. Customers are charged only when an AI agent autonomously resolves an issue from start to finish. If the customer asks to speak with a human agent or isn't satisfied with the outcome, no fee applies. Salesforce also said there are no extra charges for using Data 360 or Agentforce during AI-to-customer interactions. The model directly ties cost to successful, independent resolutions, which the company says helps organizations improve the return on investment for AI deployments.
Operational experience from Salesforce's own help portal
The product draws on operational lessons from running Agentforce on Salesforce's customer support website. The company reported that Agentforce has handled 4.3 million customer inquiries there, with 70% of cases resolved autonomously by AI agents. Alongside the new agent, Salesforce introduced the Agentforce Customer Service Portal, which lets organizations build self-service experiences where customers submit inquiries, receive personalized responses, and complete follow-up actions.
While many support teams are exploring AI for customer support, disconnected systems have often prevented meaningful outcomes. Agentforce Help Agent aims to reduce that friction by bringing everything together within a single environment.
Why this matters for customer support professionals
For support teams evaluating AI agents, a pay-per-resolution model lowers the financial risk of adoption by linking costs directly to fully resolved tickets. The 70% autonomous resolution rate from Salesforce's own deployment offers a real-world benchmark, not a vendor promise. And the unified environment means less time spent switching between tools and more time focused on the customer interactions that genuinely need human judgment.
Your membership also unlocks:
