I am heading to the US, and I am expecting to hear a lot about AI. Not a little. A lot. Enough AI noise to make a hotel lobby espresso machine look understated.
That is exciting, because AI is genuinely changing service. But it is also a little dangerous, because the loudest conversation can drown out the work that customers and employees need us to do right now.
The real question for US service leaders is not, “Which AI tool should we buy next?” It is, “Which service problems are we finally brave enough to fix, and where can AI help us do it faster, safer and better?”
AI is loud. Service failure is louder.
In the last few years, every service conversation has acquired an AI soundtrack. Chatbots, copilots, agent assist, generative knowledge, speech analytics, autonomous agents. All useful. All moving quickly.
But the evidence from CX leaders is sobering. Forrester reported in 2025 that customer experience quality is still under pressure globally, and its US CX Index work points to another year of weak performance. In plain English: customers are not handing out gold stars because a brand bought an AI licence.
That does not mean AI is the wrong conversation. It means AI needs better company. It needs service design, process simplification, clean data, knowledge management, quality improvement, workforce capability and leadership courage. Yes, the glamorous stuff.
Why the US service market matters now
The US has scale that is both inspiring and unforgiving. A small service defect can become a national problem by lunchtime. A great service model can improve millions of lives just as quickly.
That is why I am interested in extending Customer Science work into the US market. The opportunity is not just commercial. It is personal. I believe service is one of the practical ways we make everyday life better. When a customer gets the answer they need, when a patient understands the next step, when a frontline employee is confident rather than exhausted, the world becomes a little less ridiculous.
And, frankly, the world has achieved quite enough ridiculousness without adding another broken portal, duplicated form or chatbot that says “I understand” while proving the opposite.
Five service challenges hiding under the AI noise
1. Service complexity is outrunning operating models. Many organisations have added channels, suppliers, systems and policies faster than they have redesigned the service around the customer. The result is avoidable effort for customers and heroic workarounds for staff.
2. Customers want speed and humanity. They are happy to use self-service when it works, but they still want a human when the issue is complex, emotional, risky or simply weird. Shep Hyken’s customer service research is useful here because it captures the tension: customers want technology, but not at the expense of choice and trust.
3. Knowledge is too fragile for the AI era. AI is only as useful as the knowledge, policies and data it can rely on. If the source material is stale, contradictory or scattered, AI politely accelerates confusion. Very efficient confusion, but confusion all the same.
4. Leaders are measuring activity instead of improvement. Dashboards can show more contacts, faster responses and higher containment while customers still feel abandoned. Real service improvement links insight to action: coaching, process fixes, content updates, queue design, policy change and measurable outcomes.
5. Employees are carrying the gap. When tools do not talk, knowledge is hard to find and policies are unclear, frontline people become the integration layer. That is not a career path. That is a cry for architecture.
What US service leaders can do now
Simplify before you automate. Customer Science has written about this before in Automate After You Simplify. Automation should accelerate value, not laminate waste.
Design journeys around resolution. Map the moments where customers get stuck, repeat themselves, wait, abandon, complain or escalate. Then fix the causes, not just the contact volume.
Treat knowledge as operational infrastructure. The service organisation needs living knowledge: accurate, governed, searchable and updated from real interactions. This is where tools like Knowledge Quest can turn customer conversations into usable, trusted content.
Embed quality into the operating rhythm. AI quality assurance can help assess more interactions, but the value comes from the improvement loop: define, assess, improve and prove. The goal is not more scoring. It is fewer repeat problems.
Give people better systems, not just better slogans. Employees need clear guidance, useful automation, integrated desktops, meaningful coaching and permission to solve the problem. A poster about empathy is nice. A working knowledge base is nicer.
Use follow-the-sun delivery where it creates momentum. An Australia-US service model can keep analysis, design, automation, knowledge work and support moving while the other side of the world sleeps. That can shorten cycle times without turning one local team into a 24-hour emergency department.
How Customer Science can help the US market
Customer Science brings a practical mix of CX consulting, research, service design, operating-model improvement, automation, data and technology products. That combination matters because service problems rarely live in one department.
A poor customer experience might begin with unclear policy, move through a confusing digital journey, arrive in a contact centre, expose a training gap, then finish inside a dashboard no one has time to read. That is why we work across people, process, technology, management and data.
For US organisations, the unique value is not just that we can advise. It is that we can help build, embed and accelerate the change. Customer Science Insights can connect and transform real-time service data. Knowledge Quest can improve knowledge quality and publishing. CommScore.AI can help strengthen customer communications. Automation and AI services can remove repetitive work and improve speed. Consulting and operating model work can make the improvement stick.
In short: we do not want to add more theatre to the AI stage. We want to improve the service performance customers actually feel.
The benefit is better service, not shinier technology
The best AI-enabled service model is not a bot bolted onto a broken process. It is a service system where customers get clear answers, employees have confidence, leaders see the real causes of demand and improvement work happens continuously.
That is where embedded technology matters. Reporting, knowledge, QA, communication scoring, automation and operating routines should reinforce each other. If they sit in separate corners, they become another set of tabs for staff to juggle. And nobody has ever said, “Great news, I have eleven tabs open and no idea which one is telling the truth.”
The US market can gain from this approach because scale rewards disciplined design. Improve the pattern once, then replicate it. Fix the knowledge loop once, then watch thousands of interactions improve. Shorten speed-to-competency once, then give every new team member a better start.
The personal bit: service should make life better
This is the part I care about most. Customer experience is sometimes treated as a metric, a department or a slide in the annual strategy deck. It is more than that.
Service is where people meet organisations when something matters. Sometimes it is simple: a bill, a booking, a password, a delivery. Sometimes it is deeply human: health, care, family, money, stress, safety. Better service lowers effort. It reduces anxiety. It helps employees feel competent and proud. It gives leaders a way to grow without grinding people down.
That is why I am excited about the US. A larger population means larger service challenges, but also larger impact. If we can help organisations build world-leading services there, the benefit is not abstract. It shows up in millions of everyday moments where things simply work better.
A practical next step for US service leaders
If you are a US service, CX, contact centre or digital leader, start with one honest question: where are customers and employees paying the price for complexity?
Then pick one journey, one product line, one service channel or one operational pain point. Diagnose it properly. Simplify it. Improve the knowledge. Instrument the flow. Add automation where it helps. Measure the outcome. Repeat.
That is not as loud as an AI keynote. It is also much more likely to work.
FAQ: US service challenges, AI and Customer Science
Why is Customer Science looking at the US market now?
US service organisations are investing heavily in AI while still wrestling with service complexity, legacy processes, fragmented data, workforce pressure and rising customer expectations. Customer Science can help turn ambition into measurable service improvement.
Is this article saying AI is bad for customer service?
No. The point is that AI works best when it is embedded into a well-designed service model with good knowledge, clean data, clear escalation, accountable humans and measured outcomes.
What service problems should US leaders fix before automating?
Start with the journeys, policies, knowledge gaps, contact drivers, handoffs, queue rules, quality loops and employee tools that create avoidable effort for customers and staff.
How can follow-the-sun delivery help US service transformation?
An Australia-US operating rhythm can extend the working day for design, analytics, automation, knowledge management and support without asking one local team to carry every urgent task in real time.
What makes Customer Science different?
Customer Science combines CX consulting, service design, operating-model improvement, automation, data, knowledge and practical technology products. The benefit is not another tool in the stack. It is a service improvement system that can be designed, built and embedded.
Research and references
This article draws on current CX, service, AI and Customer Science material. Key references include:
- Forrester Global CX Index Rankings, 2025
- Forrester: Average US CX Index Performance Falls For Fourth Consecutive Year
- Salesforce State of Service, 6th Edition
- Qualtrics 2026 Consumer Experience Trends
- Zendesk 2026 CX Trends: Contextual Intelligence
- Shep Hyken Customer Service and CX Research
- Springer Nature: Between artificial intelligence and customer experience
- Zhang and Narayandas: Engaging Customers with AI in Online Chats
- arXiv: Agentic AI and Human-in-the-Loop Interventions
- Customer Science: Customer Experience, Service and Digital Transformation
- Customer Science: Automate After You Simplify
- Customer Science: Automating QA in Contact Centres with AI
- Unsplash image by Lx1