Salesforce
United States, Global Enterprise software
AI agents execute multi-step work Role Redesign
Deep dive available
Salesforce
Salesforce provides an unusually complete public view of how an AI deployment can change both work and workforce structure. Agentforce absorbed a large share of routine support interactions while human work shifted toward complex, proactive and consultative activity.
Read case studyFor the concepts behind task, workflow and role changes, use the AI workforce transformation knowledge guide. For individual capability development, explore the AI Fluency framework.
What changed
Routine customer-support conversations increasingly move to Agentforce, while human support engineers concentrate on more complex or proactive customer work and hundreds have moved into other functions.
Key results
Customer conversations handled by Agentforce — Agentforce had handled 2.6 million customer conversations — After 2600000 conversations
by 2026-04
Agentforce resolution rate — Agentforce reported a 63% resolution rate — After 63 percent
by 2026-04
Customer satisfaction relative to human agents — Customer satisfaction was reported as matching human agents
by 2026-04
How the organization responded
Organizational response: Redeploy · Reskill · Reduce
Why it matters
Salesforce directly evidences redeployment, reskilling and lower support staffing requirements linked to the agentic support model; the sources do not resolve how the headcount change divides among redeployment, attrition, non-backfill and other mechanisms. Provides a rare case where workflow change, actual redeployment and AI-linked reduced staffing requirements are all directly reported, while the mechanism of the headcount reduction remains partly unresolved.
Capability implications: AI Fluency
Human role / decision rights
Human escalation is evidenced: more difficult or relationship-intensive support work remains with human staff. Formal approval, veto and override rights are Not Yet Evidenced.
Source
Salesforce
First-party source documents scaled Agentforce customer-support use, a 63% resolution rate, customer satisfaction matching human agents, hundreds of support-engineer redeployments, training for role transitions and human escalation for more complex or relationship-intensive work.
The Logan Bartlett Show
CEO interview directly reports support headcount moving from about 9,000 to 5,000 and links the lower staffing requirement to the agentic support deployment. The source does not resolve the mix of redeployment, attrition, non-backfill and other mechanisms.
