AI Isn't Killing Support—It's Killing Excuses
DCX Links | March 1, 2026
Welcome to the DCX weekly roundup of customer experience insights!
Automation is doing its job. Queues are shrinking. Response times are improving.
Now comes the harder part.
This week’s stories make one thing clear: AI is shifting customer service from speed to system quality. When repetitive work disappears, what customers evaluate next is your design—handoffs, tone, consistency, effort, and whether the issue is actually resolved.
And here’s the tension:
Customers don’t want AI-only support. But they do want AI-enabled experiences that are faster, smarter, and lower effort.
The differentiator isn’t automation.
It’s how well you orchestrate it.
Let’s dig in.
This week’s must-read links:
Intercom’s 2026 Report: AI Is Moving From Efficiency to Experience
Stop “Deflecting” Customers. Start Solving Outcomes.
Chat won’t replace your UI. It should sit on top.
AI Agents Just Arrived at Your Restaurant
DCX Stat of the Week: Half Your Customers Would Walk Away From AI-Only Support
DCX Case Study of the Week: Capita Helps Financial Services Client Elevate Customer Service with AI-Powered Call Analysis
I’m obsessed with Wispr Flow Pro! Get a Free Month Now!
Intercom’s 2026 Report: AI Is Moving From Efficiency to Experience
In Intercom’s 2026 Customer Service Transformation Report, the VP of Customer Support Declan Ivory’s message is when AI handles the repetitive work, it frees up capacity. But the real shift is what comes next. Once the queue pressure drops, the focus moves to system quality. Handoffs. Tone. Channel consistency. Proactive outreach. Support stops being reactive and starts looking like a product with goals, metrics, and iteration.
The quiet message: you no longer get to blame backlog for a mediocre experience.
Where this lands for CX leaders
Your success metrics have to mature. Faster replies are expected. Customers judge you on whether their issue is actually solved without friction.
Consistency becomes competitive. It’s not about optimizing chat or email. It’s about the experience feeling the same everywhere.
Proactive becomes a differentiator. AI creates the space to anticipate needs instead of waiting for tickets.
Experience design becomes your leverage. If AI works, system quality becomes the variable you control.
The headline isn’t efficiency. It’s accountability.
🔗 Go Deeper: 2026 Customer Service Transformation Report
Stop “Deflecting” Customers. Start Solving Outcomes.
Matt Price, CEO of Crescendo AI, pushes back on the language of “containment.” If your automation strategy is built to reduce human contact, customers feel it. Crescendo only charges when an issue is resolved and won’t bill for negative experiences. That forces alignment around outcome, not volume.
The deeper point: humans aren’t the safety net. They’re part of a continuous improvement loop that trains and strengthens the AI.
The operational takeaway
Automation philosophy shows up in the experience. AI can either build trust or erode it quietly.
Escalation moments define loyalty. When AI hands off, the quality of that transition matters more than the initial response.
Outcome clarity is essential. If you can’t define what “resolved” means, you can’t optimize it.
Roles will shift, not disappear. Frontline and QA teams move toward tuning, insight, and proactive fixes.
Deploy AI as a wall and you cut cost while trust leaks out. Deploy it as a coordinated system and you improve both economics and experience.
Chat won’t replace your UI. It should sit on top.
It’s not really about hating chat interfaces.
It’s about bandwidth and effort.
Julian Lehr, Creative Director at Linear, is making a deeper point: natural language is a low-efficiency input mechanism for many computing tasks. We romanticize it because it feels “natural.” But when speed matters, we abandon language and switch to compressed signals. Buttons. Shortcuts. Gestures. Muscle memory.
His core argument:
Natural language is great for thinking, exploring, and high-fidelity exchange.
It’s bad for executing known actions quickly.
Productivity tools win because they compress intent into fast, repeatable inputs.
Conversational UI is unlikely to replace GUIs because it’s slower at task execution.
The real opportunity is augmentation: an always-on AI layer that sits across tools and increases bandwidth, not replaces them.
Underneath all of that is a more philosophical thesis:
Stop treating AI as a replacement paradigm.
Start thinking about it as a complement that expands what’s possible.
In CX terms, this is about friction vs. flow. Customers don’t care what’s “natural.” They care what’s faster, easier, and less effortful.
It’s also a quiet warning: if we force customers to “describe” what they want in chat when a single tap would do, we’re moving backward, not forward.
The butter metaphor is the tell. The future isn’t more conversation. It’s less need for it.
🔗 Go Deeper: Julian: “The case against conversational interfaces.”
AI Agents Just Arrived at Your Restaurant
Adam Brotman, co-CEO of Forum3, is saying: restaurants are about to get a new kind of workforce, and it’s going to change everything.
Most restaurant leaders still think AI is a chatbot that writes emails or takes drive-thru orders. Meanwhile, the tech has moved way past that. The gap between what AI can actually do and what operators think it can do is getting dangerous.
Here’s what he’s really arguing:
First, AI “agents” just showed up. Not bots. Not helpers. Digital workers that can analyze data, spot problems, suggest fixes, and even take action.
Second, restaurants are sitting on tons of data. POS. Loyalty. Labor. Supply chain. Reviews. Pricing. Delivery apps. It’s all there. But it’s scattered. Lots of ingredients, no chef. AI becomes the thing that connects it and turns it into insight.
Third, this isn’t about replacing hospitality. It’s about upgrading decision-making. Smarter schedules. Better pricing. Faster marketing tests. Cleaner margins.
The deeper message? Restaurants have always relied on vendors for tech. Now they can own the intelligence layer. That’s new.
Bottom line: this is less about chatbots and more about power. The operators who treat AI like a real operating shift will move ahead. The ones who don’t will slowly fall behind.
🔗 Go Deeper: Modern Restaurant Management
DCX Stat of the Week: Half Your Customers Would Walk Away From AI-Only Support
DCX Stat: 50% of consumers say they would cancel a service if they discovered its customer support was solely AI-driven, and 42% would pay extra for access to human representatives.
Takeaway: Going “AI-only” isn’t just a CX risk, it’s a churn and pricing risk. If you’re scaling automation, you still need a clear promise of human access—and you can likely justify premium tiers or fees for guaranteed human support.
Source: SurveyMonkey, “Customer service trends & statistics for 2026: Why consumers still trust humans over AI”
🔗 MORE STATS: Daily Stats on Substack Notes
DCX Case Study of the Week
Capita Helps Financial Services Client Elevate Customer Service with AI-Powered Call Analysis
CX Challenge:
A large financial services organisation was grappling with recurring customer enquiries at scale. Manual call review and QA were costly, inconsistent, and too slow to keep up with volumes — leaving contact centre teams without timely insights into issues, agent performance, and customer experience trends.
Action Taken:
The client deployed CallSight, Capita’s AI-powered analytics and insights platform. Using generative and machine learning techniques, CallSight automatically analyses every recorded call, extracts themes and patterns, and performs continuous quality assurance — replacing periodic sampling with daily, organisation-wide insight.
Result:
While specific client KPIs weren’t fully published, typical results from CallSight deployments include measurable operational uplifts:
• ~12–15% reduction in average handling time
• ~15% improvement in first-call resolution
• ~15% drop in post-call administrative work
These impacts help lower customer effort, speed resolution, and support more effective coaching.
Three Lessons for Practitioners
1. Shift from sampling to continuous insight.
Traditional QA reviews a small sample of calls. When you automate analysis across 100 % of interactions, you uncover trend drivers, root causes of dissatisfaction, and coach at scale. This elevates QA from a compliance task to a strategic lever for coaching and service improvement.
2. Use AI to augment human roles, not replace them.
Platforms like CallSight extract structured insights so humans can focus on interpretation, escalation patterns, and high‑value coaching. The operational burden shifts from manual audit to analytical action — differentiating routine patterns from true service opportunities.
3. Tie operational gains to customer outcomes.
Lower handle time and faster resolution aren’t just efficiency wins. They reduce effort for customers, shorten queues, and improve perceptions of responsiveness — outcomes that link directly to CSAT, retention, and regulatory compliance in financial services.
🔗 Further Reading: Elevating customer service with AI-powered call analysis – Capita News & Insights
I hope you enjoyed this week’s newsletter. If so, giver it a heart, restack or share it around.
See you next week.
✉️ Join 1,500+ CX leaders who stay ahead of the next customer curve.
Human-centered insights. Plug-and-play frameworks. Smart tools that actually work.
All designed for CX pros who want to build with purpose—and deliver with impact.
👉 Subscribe today and get the tools to elevate your strategy (and your sanity).