Share with your CMO
Most CX teams were built to solve problems that no longer define the job, and AI is exposing that structural lag fast. Writing at CMSWire, Trish Donnelly argues that rebooting the CX operating model means preserving four commitments, understanding customer intent, reducing friction, keeping promises, and owning failures, while rebuilding everything else around AI-enabled delivery. The practical prescription is tight: pilot one high-stakes journey like billing resolution, measure task completion and repeat contacts first, then expand. Decision rights and an executive sponsor have to move with the org chart, or the restructure is cosmetic.
What this means for your business
The CMOs most exposed here are the ones who reorganized CX reporting in the last two years but left authority with whoever held it before. A new box on the org chart, without budget control and escalation power attached to it, produces a team that can diagnose customer failures but cannot fix them. If your CX lead has to negotiate with product or ops every time AI surfaces a recurring break in the journey, the structure is already the bottleneck, regardless of how good the tooling is.
The piece makes a sharper claim underneath the pop-culture framing. AI hasn’t changed what customers judge; it has changed who is accountable for the outcome. When an AI system researches a billing error, applies a fix, and closes the loop with the customer, the experience still lives or dies on whether the promise was kept. That means the CX team needs fluency in evaluating AI outputs, not just designing journeys, and most teams haven’t been hired or trained for that. Donnelly flags validation and systems thinking as the emerging skill gap, which is the right diagnosis even if the piece, written from a practitioner perspective that favors incremental adoption, undersells how fast that gap is widening.
The piloting framework is the most actionable part of the argument, and it reframes a budget decision many CMOs are already sitting on. If you’re weighing whether to expand AI-assisted service to additional channels or journeys, the pilot metrics Donnelly names, task completion rate, repeat contact rate, and human handoff frequency, are exactly the leading indicators that tell you whether the model is holding before you scale costs and risk. The renewal or expansion call on your current CX AI vendor looks different if you’re tracking those numbers versus relying on CSAT alone.
Based on reporting from 90210 Reboot? Nah. CX Reboot? The Time Is Now in the Age of AI, originally published 2026-10-22 17:06:00.
