Why Agentic and Conversational AI Products Are Not What CX Leaders Think - with Baker Johnson of UJET (33 min)
ai-driven-innovation-economy ai-in-everyday-life ai-in-workforce-disruption ai-literacy-public-awareness
- Release date: 2026-01-05
- Listen on Spotify: Open episode
- Episode description:
Today's guest is Baker Johnson, Chief Business Officer at UJET. UJET is a next-generation cloud contact center platform that leverages AI to modernize the customer experience. Baker joins Emerj Editorial Director Matthew DeMello to explain why conversational and agentic AI often fall short in CX, and how legacy processes and fragmented data prevent meaningful results. Baker also offers practical steps for improving ROI — from redesigning workflows before automating them to aligning real-time interaction data with systems of record and building a healthier balance between human and AI agents. Want to share your AI adoption story with executive peers? Click emerj.com/expert2 for more information and to be a potential future guest on the 'AI in Business' podcast! This episode is sponsored by UJET.
Summary
- 🔄 Redesign Before Automating: Enterprises fail by paving legacy ‘cow paths’ with AI; start with a clean-sheet process overhaul to avoid amplifying inefficiencies.
- 📊 Reconcile Data Systems: Bridging systems of record and real-time interactions is crucial for accurate, accountable agentic AI in CX.
- ⚖️ Human-AI Collaboration: Reject zero-sum replacement myths; design workflows where AI and humans complement each other for optimal outcomes.
- 🎯 Outcome Over Deflection: Shift metrics from blocking customer access to delivering frictionless results, as delight is often unnecessary.
- 📝 Start Manual, Scale Smart: Begin with pen-and-paper processes to ingrain human oversight before layering AI, preventing blind trust in tools.
Insights
Why have enterprises invested billions in conversational AI without delivering expected CX gains?
Time: 0:35 – 6:19
Category: AI-Driven Innovation EconomyAnswer: Decades of treating customer experience as a cost center led to siloed processes, deflection-focused metrics, and automating legacy workflows, accelerating problems rather than solving them. True transformation requires reimagining processes from a clean sheet before deploying AI. (Start at 0:35)
Is ‘delighting’ customers the wrong goal for AI-powered CX?
Time: 7:52 – 13:10
Category: AI in Everyday LifeAnswer: Customers don’t seek delightful journeys; they want quick, frictionless outcomes like fast resolutions without unnecessary interactions. The best CX is often the interaction that never happens, shifting focus from deflection to results. (Start at 7:52)
What happens when you ‘pave the cow paths’ with AI in contact centers?
Time: 11:06 – 11:51
Category: AI in Workforce DisruptionAnswer: Applying AI to outdated, inefficient processes just makes bad experiences faster and more scalable, leaving frustrated customers and overwhelmed human agents to handle complex escalations. Leaders must redesign workflows holistically for real ROI. (Start at 11:06)
Why is the all-or-nothing myth about AI replacing humans misleading?
Time: 16:49 – 17:50
Category: AI in Workforce DisruptionAnswer: AI won’t fully replace or never replace agents; success lies in collaborative roles where humans handle empathy and complexity, augmented by AI for routine tasks. Zero-sum thinking ignores hybrid potential for better outcomes. (Start at 16:49)
How does data latency between systems doom agentic AI?
Time: 19:21 – 20:15
Category: AI-Driven Innovation EconomyAnswer: Legacy systems of record (CRM/ERP) clash with real-time interaction data, causing AI agents to fail due to poor context and accuracy. Reconciling these is essential for accountable, effective AI deployment. (Start at 19:21)
What’s next after human-AI collaboration: agent-to-agent CX?
Time: 21:46 – 22:18
Category: AI in Everyday LifeAnswer: Consumer AI agents will handle interactions directly with business agents, demanding advanced data governance and protocols like MCP. This shifts CX from human calls to seamless A2A communication. (Start at 21:46)
Should AI adoption start with pen and paper, not shiny tools?
Time: 25:02 – 26:43
Category: AI Literacy & Public AwarenessAnswer: Leading with technology skips critical human-in-the-loop learning, anomaly detection, and process understanding. Exhaust manual methods first to build intentionality, ensuring AI amplifies human strengths rather than creating blind dependency. (Start at 25:02)