Customer Experience · November 16, 2026
CX Maturity Models Compared: Why the Score Isn't the Point
A practitioner's comparison of stage-ladder, competency-scorecard and benchmark CX maturity models — and why most organisations misread their own score.
Every CX leader has sat through the same ritual: a consultant unveils a five-stage maturity ladder, the organisation places itself somewhere in the hopeful middle, and six months later nothing about the customer experience has actually changed. The model wasn't wrong. It was treated as a badge instead of a diagnostic.
A CX maturity model is a structured framework that assesses how systematically an organisation designs, governs, measures and improves customer experience — typically scoring it against defined stages or competencies rather than against competitors. The honest answer to "which one should we use" is: the model matters less than whether you use it to change how decisions get made. Most don't, because they're built to produce a score, not a mechanism.
I've run maturity workshops across banking, retail and telecoms in this region, and the pattern repeats: the assessment becomes a slide, the slide becomes a benchmark against "where we want to be by 2028," and the underlying operating model — who owns a journey, who approves a fix, who gets paged when a touchpoint breaks — stays exactly as fragmented as before. This piece compares the main families of CX maturity model on the market, explains why organisations consistently misjudge their own score, and sets out what actually needs to change underneath the score for maturity to mean anything.
Where did CX maturity models actually come from?
They're borrowed architecture. The staged-ladder format that dominates CX maturity assessment today descends directly from the Capability Maturity Model, developed by the Software Engineering Institute at Carnegie Mellon University in the early 1990s to assess software development processes. Its five levels — moving from ad hoc and chaotic through to repeatable, defined, managed, and finally optimising — were designed to answer a narrow engineering question: can this organisation reliably repeat its own good work, or does every project start from zero?
CX consultancies adopted the same scaffolding because the underlying problem is identical. Most organisations don't lack good customer moments — they lack the ability to repeat them deliberately. A five-star branch interaction or a brilliant support call is a lucky instance unless the behaviour behind it has been documented, trained, and built into the process. Maturity, in both software and service, is really a question of repeatability, not talent.
What are the main types of CX maturity model?
Strip away the branding and most commercial CX maturity models fall into one of three archetypes. Knowing which one you've been handed changes how you should read the result.
- Stage-ladder models. Direct descendants of CMMI. Organisations are placed on a single linear scale — typically four to six stages, from "reactive" or "unaware" through to "customer-obsessed" or "optimising." Simple to communicate to a board, easy to set a target against. The weakness: a single score flattens an organisation that might be excellent at complaint resolution and dreadful at onboarding into one misleading number.
- Competency-scorecard models. Rather than one ladder, these score an organisation across several independent dimensions — commonly leadership commitment, customer insight, journey design, employee experience, governance, and measurement discipline — each with its own maturity rating. The research firm Forrester has long championed this multidimensional approach in its CX practice, and Qualtrics' XM Institute publishes comparable experience-management maturity research built around parallel competencies rather than a single ladder. The advantage is diagnostic precision: you can be a 4 on leadership and a 1 on governance, and the model shows you exactly where the gap sits.
- Index or benchmark models. These don't measure internal practice at all — they measure outcomes, usually through customer survey data, and rank organisations against competitors. Forrester's annual Customer Experience Index is the best-known example: it benchmarks brands on the customer's perceived quality of experience rather than auditing the organisation's internal processes. Useful for external positioning and board narrative; close to useless as a guide for what to fix internally, because it tells you the score moved without telling you which lever moved it.
None of these is "the" correct model. They answer different questions. A stage-ladder tells the board a simple story. A competency scorecard tells an operating team where to intervene. A benchmark index tells the market how you're perceived. The mistake is using a benchmark index to make an internal resourcing decision, or using a stage-ladder's single number to brief a transformation programme — both happen constantly, and both produce false confidence.
Why do organisations consistently misjudge their own maturity score?
Because self-assessment is not a neutral act — it's a decision made under the same biases that distort every other judgment about one's own competence. Two behavioural mechanisms do most of the damage.
The first is loss aversion, the principle described by Daniel Kahneman and Amos Tversky in their 1979 prospect theory work (published in Econometrica) showing that losses register roughly twice as painfully as equivalent gains feel good. Applied to a maturity assessment, a CX team facing a self-rating exercise isn't weighing "where are we, honestly" — it's weighing "what does it cost us if leadership sees a low score." Teams round up. A function that is genuinely reactive and ad hoc will self-report as "developing," because the perceived loss of looking behind peers outweighs the honest gain of an accurate baseline. I've watched the same organisation score itself a 3.5 on "customer insight" in a self-assessment and a 1.8 when an external team scored the identical evidence a month later. The evidence didn't change. The incentive to protect status did.
The second is the goal-gradient effect — the tendency to accelerate effort as a goal feels closer, first documented by Clark Hull in the 1930s and demonstrated in a well-known consumer-behaviour context by Ran Kivetz, Oleg Urminsky and Yuhuang Zheng in their 2006 study on loyalty-card completion, published in the Journal of Marketing Research, which found customers who received a head start toward a reward purchased more frequently as they approached the goal. Staged maturity ladders exploit this deliberately: telling an organisation it's "one stage from optimised" is motivating in exactly the way a loyalty card with two stamps already filled is motivating. That's a legitimate behavioural lever when used honestly to drive momentum. It becomes a problem when it's used to flatter — placing an organisation at "stage 3 of 5" regardless of evidence, purely because "almost there" sells better than "just starting."
A maturity score that makes everyone comfortable has already failed at its one job.
The practical fix isn't a better survey instrument. It's separating the people who score the organisation from the people whose budget depends on the result, and anchoring every rating to observable evidence — a documented process, a named owner, a measured outcome — rather than a team's confidence in its own effort.
What separates a real maturity score from a cosmetic one?
This is where most commercial maturity models stop short, and where the service design lens earns its keep. A competency scorecard can tell you an organisation is "developing" on journey design. It rarely tells you why — and the why is almost always an absence of operating infrastructure rather than an absence of good intentions.
In practice, I check four things that most off-the-shelf maturity questionnaires don't ask directly, because they're structural rather than attitudinal:
- Does a service blueprint exist for the journey, and is it current? Not a journey map — a blueprint that documents frontstage actions against the backstage systems, policies and people producing them. An organisation can score well on "customer insight" and still have no blueprint, which means nobody can see where a frontline promise depends on a backend process that was never designed to keep it.
- Is there a named owner for each journey, with authority to act across departments? Maturity models love to ask about "cross-functional collaboration." The honest proxy is simpler: when a journey breaks, is there one name accountable for fixing it, or does it require a committee?
- Does a touchpoint failure trigger a defined response, or a scramble? This is the difference between a governance structure and a goodwill structure. Mature organisations have an escalation path built before the crisis, not improvised during it.
- Is customer feedback routed to the person who can change the process, not just the person who reads the dashboard? Plenty of "mature" organisations by competency-scorecard standards have excellent listening infrastructure and no mechanism connecting what they hear to what they build next.
A model that scores sentiment, intention, and leadership commitment without auditing these four structural questions is measuring enthusiasm, not maturity. Enthusiasm is necessary. It isn't sufficient, and it doesn't survive a leadership change.
How should you actually run a CX maturity assessment?
The sequence matters more than the instrument. Here's the process I use when a maturity assessment needs to produce a usable roadmap rather than a framed certificate:
- Separate the scorers from the scored. Whoever owns the budget for the improvement programme should not be the one filling in the self-assessment. Bring in an external facilitator, or at minimum a function with no stake in the outcome, to keep loss aversion out of the scoring.
- Pick the archetype that matches your decision, not your ambition. If the output needs to brief an investment case to the board, a benchmark index gives you external credibility. If it needs to direct next quarter's work, a competency scorecard gives you the granularity. Don't use one to do the other's job.
- Anchor every rating to artefact evidence. A score of "defined" on journey design should require someone to produce the actual blueprint, not describe an intention to build one. If the evidence doesn't exist, the score can't either, regardless of how confident the team feels.
- Map the score against live journeys, not the organisation in the abstract. An organisation can be mature in onboarding and immature in complaints handling simultaneously. Run the assessment per journey where you can, and aggregate afterwards — never start at the aggregate.
- Convert every gap into an owner and a date, inside the same workshop. A maturity assessment that ends with a report and no roadmap is a maturity assessment that will be repeated, unchanged, in eighteen months. Translate each structural gap directly into a governance fix: a named journey owner, an escalation path, a feedback routing change.
- Re-score on a fixed cycle against the same evidence standard. Annual re-assessment only works if the bar for evidence hasn't quietly moved. Keep the rubric identical year over year, or the "progress" you report is noise dressed as improvement.
Step three is the one organisations skip under time pressure, and it's the one that determines whether the entire exercise was honest. An assessment built on confidence rather than artefacts will always drift upward over time, regardless of whether anything on the ground has actually changed.
This structural approach overlaps heavily with quality frameworks outside the CX industry proper — the EFQM model's treatment of customer results runs on a similar enabler-and-result logic, scoring the mechanisms that produce an outcome rather than the outcome's self-reported sentiment. It's worth reading alongside any CX-specific maturity tool, because it forces the same discipline: don't score the feeling, score the system producing the feeling.
What does a mature CX organisation actually look like in practice?
Not a company with a high survey score. A company where the next good customer moment doesn't depend on which agent picks up the phone, because the process behind it has been designed, documented and owned deliberately — and where a bad moment triggers a known response instead of an improvised apology. Maturity, properly measured, is the gap between "we got lucky" and "we built it that way on purpose."
The organisations that treat maturity assessment as governance infrastructure — not a scorecard to frame — are the ones whose score actually moves between cycles. Everyone else just gets better at describing the gap they haven't closed.
If your last assessment produced a number but no change in who owns what, the model wasn't the problem. A structured CX maturity assessment paired with real governance redesign closes that gap; Renascence's own CX maturity diagnostic is a useful starting point if you want a first, evidence-anchored read before committing to a full programme. From there, the work that actually moves the needle — customer experience strategy, CX governance design, and the service design discipline that turns a blueprint into a working operating model — is where the score stops being a slide and starts being true.
FAQ
Questions we get on this topic
Related reading
Writing on how human behavior shapes the experiences brands deliver — at the intersection of behavioral economics and customer experience.
Stay ahead of CX
Get the Journal in your inbox.
Insights, frameworks and event round-ups from the Renascence team. No spam, ever.






