The New CX Operating Model for the AI Era

Over the past decade, customer experience teams have invested heavily in CRM systems, contact centre technology, journey mapping, voice of customer programmes and digital channels. AI is now changing the rules again, and many organisations are responding by treating it as simply another tool to deploy. The bigger challenge is organisational. The operating model that worked in a human-centric world was never designed for one in which AI increasingly participates in customer interactions, decisions and service delivery. AI is not simply changing customer experiences. It is changing how customer experience organisations operate.

A CX operating model is the framework that determines how an organisation delivers customer experience. It typically covers people, processes, technology, governance, metrics and organisational structure, and its purpose is to ensure that experience is delivered consistently across the business rather than depending on the efforts of individual teams.

Most CX operating models were designed around human work. Human agents handle enquiries, human managers supervise quality and human analysts review customer feedback. AI introduces a participant the original design never anticipated. A recent Gartner survey of 321 service leaders found 91 per cent feel pressure from executives to implement AI in 2026, with priorities shifting toward customer satisfaction and self-service success rather than back-office efficiency alone. As Kim Hedlin, Director of Research in Gartner's Customer Service & Support practice, put it, “AI and human expertise must work in tandem”. For the first time, customer experience organisations must design workflows that include both humans and intelligent machines.

Customer interactions once flowed purely between employees and customers. AI now participates directly, through agent assist, AI copilots, AI agents and automated knowledge retrieval. Forrester's Wave evaluation of customer service solutions for the first quarter of 2026 argues that AI is shifting service from a reactive, manual cost centre toward a proactive, personalised layer that handles the bulk of interactions. Kate Leggett, Principal Analyst at Forrester, described the change directly: “AI addresses the majority of the work, while CSRs assist AI.” The operating model must now define how humans and AI share that work, rather than leaving the split to chance.

Traditional CX has tended to react to requests as they arrive. AI allows organisations to anticipate needs instead, through churn prediction, proactive outreach and earlier detection of service issues. McKinsey's research into customer care leaders found that the strongest performers had reversed rising inbound volumes through smarter self-service and proactive engagement, reporting markedly better experience scores than organisations still treating AI as a bolt-on tool.

AI depends on data drawn from marketing, sales, service, product and operations alike, which demands a level of cross-functional collaboration many organisations do not yet have. Rather than creating fragmentation, AI tends to expose fragmentation that already existed, making disconnected data and conflicting ownership far harder to ignore than before.

Traditional processes are largely rule-based and fixed. AI enables real-time recommendations, personalisation, adaptive workflows and intelligent routing, all of which depend on decisions being made continuously rather than through periodic reviews. The operating model that results is inherently more flexible and more dependent on live data than the one it replaces.

Many AI initiatives still begin life as technology projects, owned and run by IT. Forrester's research into enterprise AI argues that this framing is itself the constraint, since operating models built for a purely digital era assumed human-only workforces and deterministic workflows that collapse once AI agents need to act within guardrails alongside people. The organisations seeing the greatest value increasingly treat AI as a customer experience capability with shared ownership, not an IT deployment measured by uptime alone.

People and roles are evolving alongside the work itself. Gartner found that almost 80 per cent of organisations expect to transition at least some agents into new roles rather than simply reducing headcount, while Forrester predicts that around 30 per cent of enterprises will build parallel AI functions, including managers who coach AI agents, operations teams who optimise their performance and specialists who step in when AI falters. Processes need similar redesign around automation, human oversight, escalation pathways and continuous optimisation. Governance, covering accountability, transparency and risk management, becomes central to the operating model, pushing leaders to build the controls needed to manage AI risk and maintain customer trust.

Three broad approaches are visible across enterprises adapting to this shift. A centralised AI team offers consistency and stronger governance, though it can sit at some distance from operational teams. Embedded AI teams sit closer to business needs but risk duplicating effort. A centre of excellence model, combining central standards with distributed execution, is increasingly the approach many enterprises are converging on as they move beyond isolated pilots toward an enterprise-wide capability.

Many organisations still treat AI primarily as a model problem, when increasingly it is a knowledge problem. Retrieval-augmented systems, AI copilots and AI agents all depend on the quality of the knowledge they draw on, and Gartner found that 58 per cent of service leaders are now aiming to upskill agents into knowledge management specialists, recognising that AI and self-service can only be as accurate as the content that feeds them. In the AI era, knowledge management is becoming customer experience infrastructure rather than a supporting function behind the scenes.

Several mistakes recur across organisations attempting this shift. Treating AI as a standalone project rather than organisational change is the most common, closely followed by automating processes that were already broken, since AI tends to accelerate existing inefficiency rather than fix it. Ignoring governance creates trust and compliance risks that surface later and at greater cost, while underinvesting in data and knowledge remains one of the most common obstacles to a successful deployment. Measuring technology usage rather than customer outcomes rounds out the list.

Future customer experience organisations are likely to include human employees, AI copilots and autonomous agents working together within a single operating model rather than as separate systems bolted to one another. Leaders will spend less time managing individual transactions and more time managing the systems, knowledge and outcomes that sit behind them, a trajectory worth picturing in more detail as the shift from technology ownership to genuine operating-model reinvention continues.

The future of customer experience is not simply about adopting AI. It is about redesigning the operating model around it. The organisations that thrive will be those that rethink how people, processes, technology, data and governance work together in an environment where intelligent machines are now part of the team.