The Five Phases of Customer Experience Maturity — and Why Most Companies Are Stuck in Phase Two

Our CEO Mark Bishof recently made a declaration that I think deserves more unpacking: the experience management era, as we have known it, is being rewritten. He is right. But understanding why, and what replaces it, requires understanding how we got here.

The customer experience discipline did not arrive at its current limitations by accident. It evolved through a series of distinct phases. Each one a genuine step forward at the time, each one eventually outpaced by the complexity of what enterprises actually needed. Most organizations today are running infrastructure built for an earlier phase. The world has moved on. Many CX programs have not.

Here is how I think about the arc — and where the real opportunity sits.

Phase One: The Survey Era

The first phase of enterprise experience management was defined by a single insight that was, at the time, genuinely revolutionary: you could ask customers how they felt, aggregate and track the responses, and use the data to improve the business. The Net Promoter Score, introduced in 2003, gave organizations a simple, benchmarkable metric that resonated with executives in a way that traditional market research never had.

For the better part of two decades, the survey was the experience program. You designed it, deployed it, measured the response rate, tracked the score, and reported the results. Organizations designed entire functions around this model. Careers were built on it.

And it worked…up to a point. The point was reached when organizations realized that scores, however carefully measured, did not reliably translate into the actions that would achieve key business outcomes. The survey told you something had gone wrong. It rarely told you what, where, why, or what to do about it unless you applied more context, and analyzed the language of experience, not just the scores. At the same time that survey programs were deploying at scale, customers started shifting their focus to increasingly diverse engagement platforms –– online forums, social media, location-based review sites –– which stole attention from survey platforms, and decreased survey response rates. 

“The survey told you something had gone wrong. It rarely told you what, where, why, or what to do about it.”

Phase Two: The Omnichannel Signal Era

The second phase was marked by an attempt to solve the survey’s fundamental limitation: it only captured a fraction of the customer’s experience. The solution seemed obvious — listen to more channels. Add the contact center. Add digital feedback. Add social listening. Add in-app feedback mechanisms. More signal coverage equals a better picture.

This was real progress. But it introduced a new problem: the data collected in each channel stayed in that channel. The contact center team had their data. The digital team had theirs. The CX team had survey results. Each had a partial view. Nobody had a complete understanding of the full customer journey, and insights from one channel weren’t being shared to the business functions who could resolve the matter, reduce the cost and effort of bad experiences, and increase overall customer loyalty. For instance, what negative product or digital experiences need to be fixed to reduce calls to a contact center? What in-store experiences are likely to create online chat experiences? What call center experiences escalate to complaints and trigger financial risk to the company? 

Most large enterprises are still living in Phase Two. They have invested significantly in multi-channel listening. They have dashboards — often many dashboards. They have more data than they can act on. And the signal that a customer is about to leave is invisible to the organization because the evidence is distributed across three departments that do not share a common platform or a common conversation.

This is the phase where the volume of data outpaced the organizational capacity to do anything coherent with it. And where the gap between what CX teams knew and what the business actually did about it became a source of real frustration.

“Most large enterprises are still living in Phase Two. They have more data than they can act on — and the signal that a customer is about to leave is invisible to the organization.”

Phase Three: The Connected Intelligence Era

The third phase is the one we are in the middle of right now. It is defined by the recognition that the value of experience data is not in any individual signal — it is in the connections between signals, across the customer’s full journey, mapped to the organizational context needed to act on them.

This is the phase that requires platform investment that most organizations have underestimated. Connecting a contact center transcript to a digital session to a survey response to an in-store interaction — for the same customer, in sequence, in real time — is not a data engineering problem. It is an architectural commitment that requires rethinking how the entire experience infrastructure is built.

At Medallia, this is what we mean when we talk about the Total Experience (TX) Profile — a unified view of every signal a customer or employee generates across their entire journey, connected to the organizational roles and teams that have the authority and responsibility to act on it. We processed 4.1 billion feedback signals last year, a 30 percent increase year over year, and expanded signal entity linking by 5x. Those numbers matter not because of their scale but because of what they represent: the infrastructure to see the full picture, not just a channel slice.

Phase Three is also where AI enters the story in a meaningful way. Not as an analytics add-on, but as the engine that makes sense of the full signal ecosystem — identifying patterns across channels that no human analyst could detect at scale, surfacing root causes rather than symptoms, and routing intelligence to the right person at the right moment. Over 650 of the world’s leading brands are now using Medallia’s Frontline Ready AI™ capabilities. The shift from AI as a research tool to AI as an operational capability is well underway.

Phase Four: The Continuous Action Era

Most organizations have not yet reached Phase Four. The ones that have are redefining what experience management looks like for everyone else. In order to achieve full value in this phase, organizations should connect listening posts, apply AI to infer patterns and trends, identify root causes, and assign actions and actors to make practical, impactful changes to improve customer experiences and business outcomes.  

Phase Four is defined by a single principle: experience intelligence that does not drive action is a cost center, not a competitive advantage. The shift is from programs that surface what is happening to systems that change what happens next — automatically, across the organization, at the speed of the business.

This requires three things working together. First, AI that moves beyond surfacing insights to recommending and triggering action — not just telling a manager what went wrong, but telling them what to do about it, and in some cases doing it. Second, an organizational model that connects experience intelligence to the operational systems and the people who own the metrics that actually change — the COO, not just the CXO. Third, a closed-loop architecture that measures the impact of the action taken, feeds that learning back into the system, and drives continuous improvement that compounds over time.

The organizations operating in Phase Four have done something that most CX programs never achieve: they have connected experience signals directly to financial outcomes. Not as a slide in a quarterly business review, but as a live, always-on system that identifies revenue at risk, operational cost drivers, and retention threats in real time — and coordinates the organizational response before the impact reaches the P&L.

Medallia’s Insights Assistant is an early expression of what Phase Four tooling looks like — a natural language AI agent that allows any user in the organization, from the frontline manager to the C-Suite, to query the full signal ecosystem conversationally and receive answers grounded in real data. Not a dashboard. Not a report. An answer, with a recommended next step. But the tools are only part of it. The organizations reaching Phase Four are also changing how they are structured — building cross-functional experience governance, embedding CX metrics into operational KPIs, and treating the CX practitioner not as a scorekeeper, but as a transformation driver and changemaker.

“The organizations reaching Phase Four have connected experience signals directly to financial outcomes — not in a quarterly review, but as a live system that identifies revenue at risk in real time.”

Phase Five: The Agentic Era

Phase Five is where the market is heading, faster than most organizations are planning for.

The defining characteristic of Phase Five is that the primary consumers of experience intelligence are no longer humans. They are AI agents — Salesforce workflows, ServiceNow processes, content personalization engines, and the custom agentic systems that enterprises are actively building to automate their own operations. These agents need to understand what is happening in the customer relationship and take action on it, without waiting for a human to read a report.

This shifts the role of the experience platform from a system that helps humans make decisions to infrastructure that enables machines to act. The implications are significant. The organizational hierarchy modeling, role-based access architecture, and signal routing logic that underpin today’s enterprise experience platforms need to be rebuilt for a world of machine consumers. The governance frameworks need to account for AI acting on behalf of the brand in real-time customer interactions. The trust framework — the permission that enterprises have earned to engage directly with their customers — becomes even more valuable as the stakes of that engagement increase.

At Medallia, this is the architecture we are actively building toward. We have spent 20+ years earning the trust of some of the world’s largest enterprises to manage their most consequential customer relationships. That trust, combined with an AI-native platform rebuilt for the demands of an agentic economy, is the combination that will define leadership in this market. The next generation of the platform — accelerated by new investment and a $500M commitment to AI innovation — is being designed for 70 million AI agents, not just 7 million human users.

So where are you?

The honest question for every enterprise leader reading this is: which phase are you in? And more specifically — is the phase you are in aligned to the phase your customers are living in?

Because customers do not experience your organization through the lens of your CX program structure. They experience it as a continuous journey — across every channel, in every interaction, in real time. And the gap between a Phase Two program and a Phase Four or Five experience is not invisible to them. It shows up in response times, in the feeling of being known or unknown, and in whether the organization acts like it remembers what just happened.

Most organizations know they need to evolve. The challenge is that each phase transition requires not just a technology upgrade but an organizational one. Different skills, different structures, different ways of defining success. The CX practitioner who excelled at Phase One was a survey designer. Phase Three requires a data architect. Phase Four requires a business transformation agent who sits at the intersection of experience intelligence and operational accountability. And Phase Five organizations are embracing agentic AI, workflow automation, and linking experience signals operationally to business processes and outcomes. 

The good news is that the transitions between phases are not taking as long as they used to be. The AI capabilities that took years to build are now available, at enterprise scale, as part of a platform that is already embedded in how the world’s most complex organizations run their businesses. The path from Phase Two to Phase Four is shorter than most organizations realize.

The question is not whether to make the journey. The question is whether you start now, or whether you start after your competitors already have.