Why Definitions of Customer Experience Design Keep Shifting

Most definitions of customer experience design are wrong — not because they describe something false, but because they describe something finished. They treat CX design as a discipline with settled edges, a stable vocabulary, and an agreed methodology. It has none of those things. The definition keeps shifting because the thing itself keeps shifting, and anyone who tells you otherwise is selling you a framework, not the truth.

That instability is not a problem to solve. It is the most useful signal in the field.

Understanding why the definition moves — what forces pull it in different directions, which expansions are genuine and which are scope creep dressed up as insight — is more valuable than memorising any single definition. It tells you what the discipline is actually becoming, and where the real leverage sits for organisations trying to design experiences that hold up under pressure.

What the original definition got right (and why it was always incomplete)

The earliest coherent articulation of customer experience as a designed thing came from Pine and Gilmore's 1998 Harvard Business Review piece, "Welcome to the Experience Economy". Their argument was structural: as economies mature, value shifts from commodities to goods to services to experiences, and experiences are staged, not merely delivered. The customer is the product, in a sense — their internal state is what you are designing.

That framing was genuinely original and remains partially correct. But it carried a flaw that took years to name: it treated experience as performance. The organisation stages; the customer receives. The model is theatrical, which means it is also asymmetric. The customer is an audience, not a co-author.

For a decade, CX design operated largely within that asymmetry. Journey maps were drawn from the organisation's perspective of what it thought customers experienced. Touchpoints were catalogued by channel owners. Service blueprints mapped backstage processes to frontstage interactions. All useful. All still treating the customer as the object of design rather than a participant in it.

The first definition, then, was: customer experience design is the deliberate shaping of interactions between an organisation and its customers across touchpoints, with the goal of producing a positive emotional response. Reasonable. Incomplete. And already under pressure from the moment it was written.

Why "touchpoints" became the wrong unit of analysis

The touchpoint-centric view of CX design dominated for good reason: it was tractable. You could list touchpoints, assign owners, measure satisfaction at each one, and improve them individually. It gave CX teams something concrete to manage.

The problem is that customers do not experience touchpoints. They experience journeys — sequences with memory, anticipation, and emotional momentum. Daniel Kahneman's peak-end rule, established through his research on the psychology of experienced utility, makes this precise: people's retrospective judgement of an experience is disproportionately shaped by its most intense moment (the peak) and its final moment (the end). A string of adequate touchpoints followed by a poor resolution is remembered as a poor experience. A difficult process redeemed by a genuinely warm close is remembered more kindly than the individual moments would predict.

This is not a minor refinement. It means that optimising individual touchpoints independently — the standard output of a touchpoint-centric CX programme — can actually degrade the overall experience if it flattens the emotional arc or neglects the ending. The unit of design had to expand from the touchpoint to the journey, and from the journey to the emotional arc across the journey.

That shift forced a redefinition. CX design was no longer just about making each interaction better. It was about shaping the sequence, the emotional trajectory, and the memory — which is a fundamentally different design problem. It requires different tools, different metrics, and different organisational capabilities.

The behavioural turn: when psychology became structural

The second major expansion came when behavioural economics moved from academic curiosity to operational practice. Richard Thaler and Cass Sunstein's work on choice architecture — formalised in their 2008 book Nudge — gave CX designers a vocabulary for something they had always done intuitively but rarely systematically: the structure of a choice shapes the choice made, independent of the options available.

Defaults, sequencing, framing, effort gradients — these are not decorative. They are load-bearing elements of experience design. A digital onboarding flow that requires customers to opt in to useful features will see lower adoption than one that makes those features active by default with an easy opt-out. Not because customers are irrational, but because effort is a cost, and System 1 thinking (fast, automatic, heuristic-driven) governs most low-stakes decisions. Designing against that reality is not manipulation; it is literacy.

The integration of behavioural economics into CX design changed the definition again. CX design was no longer just about the emotional quality of interactions. It was about the cognitive and motivational architecture within which customers make decisions, form expectations, and interpret what happens to them. That is a much larger canvas — and it requires practitioners who can read a journey map and a behavioural audit simultaneously.

This is where many organisations still fall short. They hire experience designers who understand aesthetics and flow, and separately hire researchers who understand behaviour. The two rarely talk to each other with enough precision. The definition has expanded; the organisational capability has not kept pace.

The employee experience correction

Around the mid-2010s, a quieter but equally important shift occurred: the recognition that customer experience design cannot be separated from employee experience. This was not a new idea — service-profit chain logic had been articulated by Heskett, Jones, Loveman, Sasser, and Schlesinger in their 1994 HBR paper — but it took time to be taken seriously as a design constraint rather than an HR aspiration.

The mechanism is direct. Frontline employees are not just delivery agents for a designed experience; they are the experience, in many moments. A customer service protocol designed without regard for the cognitive load it places on the agent, or the discretion it allows them, will produce inconsistent and often poor outcomes regardless of how well the protocol reads on paper. You cannot design the customer's experience independently of the employee's experience of delivering it.

This forced another definitional expansion. CX design now had to encompass the conditions under which employees work — their tools, their authority, their emotional labour, their sense of purpose. Service design, which had always attended to both frontstage and backstage, became more central. The definition grew to include: the design of the systems, roles, and conditions that enable employees to deliver the intended experience consistently.

That is a significant scope expansion. It means CX design touches organisational design, process architecture, technology selection, and cultural norms. It is no longer a function that sits comfortably within a single team or budget line.

Digital transformation and the fragmentation problem

Digital transformation added velocity and complexity in roughly equal measure. The proliferation of channels — app, web, WhatsApp, in-branch, call centre, self-service kiosk, social — did not just add touchpoints. It fragmented the customer's experience across contexts that are often owned by different teams, built on different systems, and measured by different metrics.

The design challenge became one of coherence under fragmentation. A customer who begins a mortgage application on a bank's mobile app, calls the contact centre to clarify a document requirement, visits a branch to submit paperwork, and then waits for a decision communicated by email is not experiencing four separate interactions. They are experiencing one journey across four channels — and the experience of that journey is determined largely by how well those channels share context, maintain consistency, and reduce the customer's effort in bridging the gaps.

This is where the concept of journey orchestration entered the CX design vocabulary. Orchestration is the active management of the customer's experience across channels and time — not just designing each channel well, but designing the transitions, the handoffs, and the continuity. It requires data infrastructure, governance, and cross-functional authority that most organisations do not yet have.

The definition expanded again: CX design is now also the governance of experience coherence across channels, time, and organisational boundaries. That is a political and structural challenge as much as a design one.

The measurement crisis and what it revealed

Every expansion of the definition created a corresponding measurement problem. If CX design encompasses emotional arcs, behavioural architecture, employee conditions, and cross-channel coherence, then NPS — a single number measuring likelihood to recommend — is not a design tool. It is a lagging indicator of aggregate sentiment, useful for tracking direction but useless for diagnosing cause or guiding specific design decisions.

CSAT and CES (Customer Effort Score) are more proximate to specific interactions, but they still measure outcomes rather than the design variables that produce them. The field has been slow to develop metrics that are genuinely diagnostic — that tell you not just that an experience was poor, but which design element failed, why, and what to change.

The Voice of Customer discipline has matured significantly, but the gap between listening and acting remains wide in most organisations. Data is collected; insight is extracted; recommendations are made; change is slow. The measurement infrastructure has not kept pace with the expanded definition of what is being designed.

This is one reason the definition keeps shifting: each new measurement capability reveals a dimension of experience that was previously invisible, which in turn reveals a design lever that was previously unmanaged. Better measurement does not just tell you how you are doing — it tells you what you are actually designing, which turns out to be more than you thought.

The current frontier: experience as a system, not a surface

The most current and contested expansion of the definition treats customer experience design as systems design — the shaping of the entire sociotechnical system within which customer value is created, not just the visible interactions at its surface.

This framing draws on service-dominant logic (Vargo and Lusch, 2004 and subsequent work), which argues that value is not embedded in products or services but co-created in use. The customer is not a recipient of value; they are a co-producer. Designing the experience means designing the conditions for that co-production — the information available, the tools provided, the social context, the feedback loops, the recovery mechanisms when things go wrong.

Under this definition, CX design is inseparable from service design, product design, organisational design, and data strategy. It is not a layer applied on top of operations; it is a lens applied to the entire organisation. That is a genuinely large claim, and it is not universally accepted — partly because it is intellectually correct and partly because it is operationally inconvenient for organisations that prefer their functions to have clean edges.

The inconvenience is the point. The definition keeps shifting because the discipline keeps discovering that experience is produced further upstream than previously thought — in decisions made in product roadmap meetings, in data governance policies, in how a company structures its complaints process, in whether a frontline employee has the authority to resolve a problem without escalation. Pulling that thread far enough leads you to the whole organisation.

What a shifting definition demands of practitioners

If the definition is genuinely unstable — not through confusion but through legitimate expansion — then the practical implication for CX leaders is clear: the discipline requires a different kind of practitioner than it did ten years ago.

The early CX designer needed empathy, visual thinking, and the ability to facilitate a journey mapping workshop. Those skills remain necessary. They are no longer sufficient.

The contemporary CX design practitioner needs:

• Behavioural fluency — the ability to read an experience through the lens of cognitive bias, effort, and decision architecture, and to design interventions that work with human psychology rather than against it.
• Systems thinking — the capacity to trace the upstream organisational causes of downstream customer outcomes, and to make the case for structural change rather than surface remediation.
• Measurement literacy — not just familiarity with NPS and CSAT, but the ability to design measurement frameworks that are diagnostic, proximate to design decisions, and connected to business outcomes.
• Cross-functional authority or influence — because experience is produced across functions, the designer who cannot work across organisational boundaries will only ever improve the parts they own, which is rarely enough.
• Commercial grounding — the ability to connect experience improvements to revenue, cost, and risk outcomes, which is the language that secures resources and sustains programmes through leadership changes.

If your organisation's approach to customer experience is still primarily a journey mapping exercise owned by a small team, with NPS as the primary success metric and no structural connection to product, technology, or people decisions, then the definition has moved on without you. That is not a criticism — it describes most organisations. It is an honest diagnosis of the gap between where the discipline is and where most practice sits.

The one thing the shifting definition confirms

Every expansion of the definition of CX design has moved in the same direction: further upstream, further inside the organisation, and further into the conditions that produce experience rather than the experience itself. The theatrical model — organisation stages, customer receives — has been progressively replaced by a systems model in which experience is an emergent property of the whole organisation's design.

That convergence is not accidental. It reflects a genuine truth about how value is created and destroyed in service relationships. You can polish every touchpoint and still produce a poor experience if the underlying system is misaligned. You can have a beautiful app and a broken promise. You can score well on individual satisfaction surveys and still lose customers at renewal because the cumulative emotional arc of the relationship never built trust.

The maturity of a CX programme is best measured not by the sophistication of its tools or the size of its team, but by how far upstream its design authority reaches. Organisations that design experience only at the surface will always be reacting to problems created deeper in the system. Those that design the system itself — the policies, the processes, the employee conditions, the data flows, the recovery mechanisms — are the ones whose customer experience holds up under pressure, at scale, and over time.

The definition keeps shifting because the discipline keeps earning the right to a larger claim. The question for any CX leader is not which definition to adopt, but how much of that larger claim their organisation is actually ready to act on. That gap — between the definition's frontier and an organisation's operational reality — is where the most consequential design work happens.

Definitions settle when disciplines stop growing. This one has not stopped yet.