USGBC Cut Case Resolution 22% Before Agentforce Even Touched a Ticket

The U.S. Green Building Council, which manages LEED certification across 180 countries, spent roughly three years under CIO Kapil Kher consolidating three disconnected knowledge platforms and two separate intake systems onto a single unified Salesforce platform. The result, published as a Salesforce case study: 30,000 cases closed year-to-date and a 22% improvement in resolution time, dropping average case resolution from 4.5 days to 3.5. Kher frames the gain in mission terms rather than efficiency alone: “I don’t look at that as an efficiency measure. I look at that as our responsiveness to our global community.” Only after that platform consolidation is USGBC now prototyping Agentforce, Salesforce’s AI agent product, to automate tier-one case triage and route tickets to the right subject-matter experts.

That sequencing — unify the data and workflow first, add the agent layer second — matches how Salesforce is now publicly categorizing what mature agentic deployment looks like. Its 2026 Customer Success Awards, going out to customers ahead of Dreamforce, group winning implementations into three pillars: “Making Data Work Smarter,” “Transforming the Way Work Gets Done,” and “AI Agents Working Together.” USGBC’s three-year consolidation is squarely the first pillar; its tier-one triage pilot is the start of the second. The awards framework reads less like a marketing taxonomy and more like an implicit maturity model — one most organizations evaluating Agentforce or any competing agent platform are still working through in the same order.

The case worth taking from USGBC isn’t the 22% number by itself — it’s that the number came from unglamorous platform consolidation, not from the AI agent Salesforce is now promoting. Any RevOps or customer-success leader sizing up an agentic rollout should treat that as the actual prerequisite: an agent triaging tickets across three disconnected systems would have less to work with than one triaging tickets after the data is already unified.