The Front Door and the Plumbing: Why Customer Experience Became the Banner Every Transformation Marches Behind
When the satisfaction score is poor, the score is not the problem to be fixed — it is the gauge reporting that something in the operating model is producing a bad experience.
Executive Summary
Somewhere in the last few years, transformation acquired a new vocabulary. Programmes that would once have been named for their systems, their processes, or their cost lines are now named for the customer. The re-platforming becomes a customer experience programme; the operating-model review becomes a journey transformation; the cost-reduction exercise arrives dressed as a promise to make things effortless. This essay is about that shift — why customer experience became the banner under which so much organisational change now marches, what genuine forces put it there, and why the banner so often outruns the change beneath it.
The argument is not that customer experience is a fad. It is that experience has become the acceptable language of transformation — the one framing that unites the boardroom, the technology function, and the marketing department around a single word — and that this very usefulness is where the danger lies. A programme justified by experience tends to be judged by the parts of experience that are visible: the redesigned screen, the shorter journey, the satisfaction score that ticks upward at launch. The parts that actually determine experience over time — the integration between channels, the state of the systems of record, the handoffs between functions, the incentives that decide whether a customer’s problem is anyone’s job — are slower, less photogenic, and easier to defer.
The result is a recognisable pattern: organisations redesign the front door and leave the plumbing untouched, then wonder why the promised transformation stalls. The essay works through why this happens, why intelligent people keep doing it, and what the strongest case for the opposite view gets right. It ends where it should: with the uncomfortable proposition that customer experience is not a thing you build but a readout of the operating model you already have — and that treating it as the deliverable, rather than the diagnostic, is the quiet reason so many experience-led transformations disappoint.
The Applause at the Launch
There is a moment, familiar to anyone who has sat through enough of these programmes, when the new digital channel goes live. The screens are clean. The eleven steps have become four. The colour palette is warm, the copy is human, the thing works on a phone. Someone demonstrates it on the big screen in the town hall and there is real, unforced applause — because it is better, visibly and immediately better than the thing it replaces. The launch survey, taken in those first weeks by the customers most willing to try something new, shows satisfaction up. A number goes on a slide. The programme has, by every measure it set for itself, succeeded.
Then the quarters pass, and something stubborn happens, or rather fails to happen. The contact centre volumes do not fall the way the business case promised. First-contact resolution sits where it always sat. Complaints migrate rather than disappear — the customer who used to wait on hold now sends a message through the shiny new channel and waits just as long for an answer that has to come from the same place it always came from. The satisfaction number, measured again a year on against the whole customer base rather than the early adopters, has drifted back toward where it began.
Nothing was faked. The front end really did change. What did not change was everything behind it, and it turns out that everything behind it was where the experience actually lived. This is the pattern the essay is about, and the first thing to say about it is that it is not a story of incompetence. The people who build the beautiful front end are good at their jobs. The pattern persists precisely because it is produced by intelligent responses to real pressures — which means it cannot be fixed by trying harder at the same thing.
Why Experience Became the Banner
To understand why experience became transformation’s favourite banner, it helps to see what problem the banner solved. For most of the previous decade, large-scale change inside organisations was justified in the language of its own machinery. You transformed because a system was end-of-life, because a regulator demanded it, because a merger had left you with three of everything, because costs had to come out. These are true reasons and they are also narrow ones. They mobilise the function that owns the problem and almost no one else. The technology director cares about the end-of-life platform; the branch network does not, until the day it breaks.
Customer experience is the first framing in a long time that crosses those boundaries. It gives the boardroom a reason to care that is not merely defensive. It gives technology a story to tell about work that would otherwise read as plumbing. It gives marketing a claim on the operating model it never previously had. And it lands at precisely the moment when the ground beneath every incumbent has genuinely moved. The rise of the smartphone as the default way people transact, the arrival of pure-digital challengers who never carried the weight of a branch estate or a mainframe, the normalisation of same-day everything by the large online retailers — these are not marketing anxieties, they are structural shifts, and they have made the quality of the digital interaction a matter of survival rather than polish.
Experience became the banner not because organisations suddenly discovered the customer, but because it was the only word large enough to unite the boardroom, the technology function, and the marketing department around a single programme — and the only word urgent enough to loosen the budget.
So the enthusiasm is earned. Something real did change in the world, and experience is a reasonable name for the thing that now has to change inside the organisation in response. The trouble is not the diagnosis. The trouble is what happens to a diagnosis once it becomes a banner — because banners are chosen for how they rally, and the things that rally a large organisation are rarely the things that are hardest to do.
The Seduction of the Surface
The surface of an experience — the interface, the journey as the customer sees it — has three properties that make it almost irresistible as the place to spend the money, and each of them is a trap.
The first is that it is visible. You can put the redesigned screen on a slide. You can demonstrate it. A senior leader can hold it in their hand and feel the improvement. Nothing in the operating model has this quality. No one has ever received applause in a town hall for retiring a piece of middleware or reconciling two customer databases into one, even though those acts do more for the lived experience than any amount of interface design.
The second is that it is fast. A capable digital team can redesign a journey in a matter of months. Re-architecting the systems behind it, untangling the integration, rebuilding the process so that a customer’s request flows through without a human rekeying it into a thirty-year-old system of record — that is measured in years, and it produces nothing to show for a long time. Under the pressure of a business case that has to demonstrate benefit within the financial year, the fast, visible work will always win the argument against the slow, invisible work, regardless of which one matters more.
The third, and the most insidious, is that the surface appears to be the whole thing. When the customer’s frustration shows up as a clunky screen, it is entirely natural to conclude that the screen is the problem. But the screen is usually the last few inches of a very long pipe, and the frustration was manufactured much further back — in the fact that the mobile channel and the branch cannot see the same version of the customer, in the handoff between sales and fulfilment where accountability quietly evaporates, in the incentive structure that rewards the contact centre for handling calls quickly rather than for making the next call unnecessary. Redesign the last few inches and you have made the pipe prettier at the end. The water arriving through it is unchanged.
What the Engine Actually Determines
Consider a composite that will be recognisable to anyone who has worked inside one of these programmes. An organisation — it could be a bank, an insurer, a utility, a telco; the shape is the same across all of them — sets out to transform the experience of opening a new account. The old online journey ran to eleven screens and asked for the same information more than once. The programme cuts it to four. It is genuinely elegant. At launch, the satisfaction score among new applicants rises by something like fifteen points, and the programme is held up as the template for everything that follows.
Two quarters later, the numbers that were supposed to move have not. First-contact resolution in the contact centre sits stubbornly around sixty per cent, exactly where it sat before. The reason is not on any screen. Behind the four beautiful steps, roughly a third of applications still fall out of the automated flow and drop into a manual queue, where a person rekeys them into a system built in the early nineties that the digital layer can talk to only in batches, overnight. The customer who falls into that third has a worse experience than they had before, because the elegant front end promised them speed and the engine behind it cannot deliver it. Their satisfaction, measured honestly, is lower than it was under the old, ugly, honest eleven screens.
“The experience the customer saw had changed. The experience the customer got had not.”
The figure that matters here is not the fifteen-point launch bump. It is the one-in-three fallout rate, and the batch window, and the sixty per cent that never moved — because those are the operating model, and the operating model is what the customer actually experiences over the life of the relationship, long after the launch survey has been filed. An experience is not something an organisation designs and then hands to the customer intact. It is something the organisation produces, continuously, out of its systems, its data, its processes, and its incentives. The interface is where that production becomes visible. It is not where it happens.
| What the programme optimised | What actually determined the experience |
|---|---|
| The number of screens in the journey | Whether the channels share one view of the customer |
| The clarity of the interface copy | Whether a request flows through without a manual rekey |
| The satisfaction score at launch | The fallout rate into manual queues two quarters later |
| The visible speed of the front end | The batch window between the digital layer and the system of record |
The Strongest Case for Doing It Anyway
It would be too easy to stop here, because there is a serious argument on the other side, and it is held by some of the most capable people in this field. The argument runs like this. Organisations that begin with the operating model never move. Start with the systems, the data, and the org chart and you will spend three years in an internal programme that the customer never sees, that loses its sponsor in the second year, and that is quietly folded into business-as-usual before it delivers anything. The genius of leading with experience is that it works outside-in. It starts from what the customer actually feels, it creates a visible, sharable artefact that sustains executive attention, and — crucially — the redesigned front end generates a pull. Once the beautiful journey exists and starts failing at the seams, the pressure to fix the plumbing behind it becomes concrete and undeniable in a way it never was when the plumbing was an abstract internal concern. You lead with the surface not because it is the whole job but because it is the only lever that moves the rest.
This is right, and it is important, and any argument that ignores it is dishonest. Outside-in genuinely does beat inside-out as a way to start. The experience genuinely is the correct place to look, because it is the one vantage point from which the failures of the operating model become visible as the customer feels them rather than as the organisation excuses them. The mistake is not leading with experience. The mistake is stopping at it.
The pull the argument describes is real, but it only does its work if leadership follows the experience down into the engine — if the launch of the front end is understood as the beginning of the diagnosis rather than the end of the programme. In practice, the pull is nearly always released too early. The applause at the launch is taken as the signal that the job is done, the sponsor moves on to the next visible win, the budget that should have funded the slow rebuild of the systems behind the interface is declared spent, and the programme books its benefits on the strength of a satisfaction bump that will not survive contact with the operating model. The outside-in argument is a case for where to start. It is routinely misread as a case for where to stop, and that misreading is the whole disease.
Experience as a Readout, Not a Deliverable
If there is a single reframe that separates the transformations that hold from the ones that drift, it is this: customer experience is not a deliverable. It is a readout.
A readout is an instrument reading. It tells you the state of the thing being measured; it is not itself the thing. When the satisfaction score is poor, the score is not the problem to be fixed — it is the gauge reporting that something in the operating model is producing a bad experience. You can, of course, tamper with the gauge. You can redesign the interface so that the launch survey reads better, in the same way you can hold a thermometer near a window to change the number without changing the temperature of the room. The reading improves. Nothing real does. And because experience is a readout that the whole organisation can now see, tampering with the gauge has become one of the most seductive failure modes available to a transformation, precisely because it looks so much like success.
Read the right way, though, the instrument is the most valuable thing the organisation has. The journey map that shows where customers struggle is not a design document; it is a fault-finding chart for the operating model. Every point of friction the customer feels is pointing at a seam behind the interface — a place where two systems do not talk, where two functions do not share accountability, where an incentive rewards the wrong thing. The great waste of the experience movement is that organisations have learned to produce these maps beautifully and then to act only on the last few inches they describe, when the map’s real value is as a set of coordinates for the far harder work upstream.
Treat the journey map as a design brief and you will redecorate the front door. Treat it as a fault-finding chart and it will tell you, seam by seam, exactly which parts of the operating model your customers are paying for.
This is why the discipline of measuring experience — the satisfaction indices, the promoter scores, the voice-of-the-customer programmes that have become standard equipment — is worth far more as diagnosis than as scorecard. Used as a scorecard, a single number rolled up to the board, it invites exactly the surface optimisation this essay has been describing; the incentive is to make the number move, and the cheapest way to make the number move is to work on the surface. Used as diagnosis, disaggregated and traced back to its causes in the machine, the same measurement becomes the thing that keeps a transformation honest, because it refuses to let the organisation confuse a better-looking front end with a better business.
What Follows From Taking This Seriously
None of this is an argument for turning inward, for the multi-year systems programme with no visible face, for making the customer wait while the organisation rebuilds its foundations in private. That approach fails for exactly the reasons the outside-in case sets out. The argument is narrower and, I think, harder: that leading with experience obliges you to keep going past the point where the applause makes it comfortable to stop.
Concretely, that means a few things that sit awkwardly against the way these programmes are usually run.
- It means writing the business case so that the visible redesign is explicitly the first phase of a diagnosis, not the deliverable — with the money for the operating-model work protected in advance, before the launch bump gives everyone permission to declare victory.
- It means measuring experience where it is produced and not only where it is seen: not the launch survey among early adopters, but the fallout rates, the repeat-contact rates, the first-contact resolution, the seams between channels — the readouts that report on the engine rather than the paint.
- It means holding the sponsor past the launch. The most reliable predictor of whether an experience transformation becomes real is whether its executive owner is still accountable for it two quarters after the celebration, when the interesting work has become invisible again.
- And it means being honest, in the room, about the difference between an experience the customer sees and an experience the customer gets — because the whole failure mode depends on the two being quietly conflated, and it survives only as long as no one insists on separating them.
The deeper lesson, the one that outlasts this particular moment of digital disruption, is about the gap between the intent of a transformation and its reality. That gap is almost never a failure of ambition or of design talent. It is a failure of following through — of mistaking the visible half of the work for the whole of it because the visible half is where the recognition lives. Customer experience did not create that gap. It has simply given it the most flattering disguise it has ever worn, because a redesigned front end looks so much more like transformation than the slow, unglamorous, decisive work of rebuilding what stands behind it.
The organisations that will still be standing when the current wave of disruption has done its work are not the ones with the most beautiful front doors. They are the ones that understood the front door was telling them something about the house — and went inside.