Most CX methodologies start with the experience: map the journey, find the friction, fix it. Ours starts one step earlier, with a structured diagnosis of what actually causes customers to choose, stay or leave. The whole process runs through six stages, from executive alignment through to a validated set of experience improvements and a proper handover. The discipline itself has a clear boundary: stages one to five diagnose and validate, they do not deliver.
What happens after that is a separate stage, one we're equally placed to support, whether that means handing validated improvements to your own delivery teams or carrying them through with our digital product, AI and service model design capability.
The six stages below describe the full arc of a Causal CX engagement, from agreeing what growth and retention means for your organisation through to a validated set of experience improvements and a considered handover.
Before any research begins, we agree what growth and retention actually means for this engagement, and which customer populations are in scope. Organisations that serve genuinely different customer bases, for example residential, business and government accounts with different jobs and different competitors, need this decided explicitly, not assumed. The outcome is a signed off programme charter: purpose, scope, a growth-based definition of success, and the populations we are focused on. Jobs to Be Done is a theory of customer choice developed by Clayton Christensen. It holds that customers don't buy products or services for their features, they hire them to make progress in a specific circumstance. Understanding the functional, emotional and social dimensions of that progress is the most reliable way to identify what actually drives customer decisions.
We turn that scope into a concrete sampling plan. Because the diagnostic is built to surface pull, anxiety and habit as well as push, the highest signal interviews come from people living through a real decision: customers who recently switched in, customers who recently switched out, near switchers who seriously considered leaving but stayed, and non-customers who chose a competitor instead. A long tenured, comfortable customer is often the weakest source of insight here, precisely because those forces have gone quiet for them. The outcome is a research sampling and recruitment plan which drives the research effort.
The qualitative research is what everything downstream is built on: in depth, switch style interviews across the sample, exploring the real decision, the alternatives considered, and the functional, emotional and social dimensions customers used to judge them. Including non-customers who chose a competitor is what sets this apart from conventional Voice of the Customer research, which only ever talks to existing customers and therefore only ever hears about push. The outcome is three artefacts:
We turn these three artefacts into a prioritised, weighted set of dimensions, the design brief for everything that follows. We weight these dimensions by the strength and consistency of the patterns across the diagnostic, not by asking customers to score or trade off Jobs statements directly. Customers do not think in the language of Jobs to Be Done. It is our analytical construct, not theirs, so asking them to rate it quantitatively risks manufacturing precision rather than finding it. The outcome is a prioritised performance framework, the criteria against which the current experience, and every proposed improvement, will be judged.
We then build a customer lifecycle model, a visualisation of how customers engage with the organisation across their whole relationship, not just within a single transaction, drawn directly from the population scoping and the diagnostic rather than as a separate exercise.
Finally we develop channel and customer typologies that capture how different types of customers pursue the same underlying jobs differently. A digital native customer and a customer with low digital confidence may be hiring an organisation to make identical progress. They are weighted by the same causal dimensions. But they need those dimensions delivered through entirely different channels. We keep this explicitly separate from the Jobs work. The jobs are shared across these customer types; the mode of pursuing them is not. The outcome is a typology, framed as channel and capability customer types rather than Jobs based segments.
We turn these three artefacts into a prioritised, weighted set of dimensions, the design brief for everything that follows. We weight these dimensions by the strength and consistency of the patterns across the diagnostic, not by asking customers to score or trade off Jobs statements directly. Customers do not think in the language of Jobs to Be Done. It is our analytical construct, not theirs, so asking them to rate it quantitatively risks manufacturing precision rather than finding it. The outcome is a prioritised performance framework, the criteria against which the current experience, and every proposed improvement, will be judged.
We then build a customer lifecycle model, a visualisation of how customers engage with the organisation across their whole relationship, not just within a single transaction, drawn directly from the population scoping and the diagnostic rather than as a separate exercise.
Finally we develop channel and customer typologies that capture how different types of customers pursue the same underlying jobs differently. A digital native customer and a customer with low digital confidence may be hiring an organisation to make identical progress. They are weighted by the same causal dimensions. But they need those dimensions delivered through entirely different channels. We keep this explicitly separate from the Jobs work. The jobs are shared across these customer types; the mode of pursuing them is not. The outcome is a typology, framed as channel and capability customer types rather than Jobs based segments.
We commit no capital to delivery until an idea has been validated with real customers. We reframe every idea as a falsifiable hypothesis tied to a named causal driver. We agree the success threshold before testing, not after. Then we run the most cost effective test that produces evidence of actual behaviour rather than stated intent: a concierge test, a wizard of oz test, a smoke test or fake door, or a small cohort pilot. Every idea gets an explicit decision: persevere, pivot or stop. The outcome is a validated, evidence backed set of experience improvements, each with a documented expected impact on a named causal driver.
The validated improvements from stage five are ready to move into delivery, either with your own teams or with ours, alongside a short set of recommendations for how your Voice of the Customer practice and measurement systems should evolve so the causal discipline outlives the engagement. In practice this means updating discussion guides to keep probing pull, anxiety and habit alongside push, and adding a small number of leading indicators to your existing measurement framework, hypothesised to predict choice rather than only describe satisfaction, tracked against actual growth outcomes over the following twelve to eighteen months. Either way, we stay engaged through the handover to make sure intent survives contact with delivery. The outcome is a transition plan, an updated measurement approach, and a clear decision on who takes the validated improvements forward.
Our approach rests on a small set of established academic and strategic frameworks. Jobs to Be Done, developed by Clayton Christensen, is our primary lens for understanding customer choice: why customers hire, stay with, or fire an organisation. The Four Forces of Progress, developed by Bob Moesta alongside Christensen, explain the competing pushes and pulls that determine whether a customer actually switches. Richard Rumelt's Good Strategy shapes how we diagnose problems and structure recommendations, so a roadmap states a genuine strategic position rather than a menu of options. Eric Ries's Lean Startup and The Startup Way inform how we test ideas against real customer behaviour before committing capital to build anything.These frameworks are not references we drop into a proposal to lend it authority. They are the thinking underneath every diagnosis, every weighting decision in the performance framework, and every decision to persevere, pivot or stop that we make in customer validation.
Elemental Concept has developed two purpose built tools that support the work.
The performance framework is a structured method for turning a Jobs and CX diagnostic into a prioritised, weighted set of causal dimensions. We weight each dimension by the strength and consistency of the patterns we find across the diagnostic interviews, not by asking customers to score or trade off Jobs statements directly. The result is a defensible, ranked view of what actually drives choice, and a clear rationale for why one improvement is prioritised ahead of another.
Our customer journey mapping application is a web based platform that keeps the customer lifecycle model, customer types and critical journey prioritisation live and current, rather than locked in a static document that goes out of date the day it's delivered. Clients can maintain and extend it themselves once the engagement ends.
Causal CX itself is a diagnostic and validation discipline. Stages one to five produce a validated, evidence backed set of experience improvements, and that boundary is deliberate: an organisation judging every recommendation by its effect on delivery, or on sentiment scores, is the surest way to slide back into conventional CX. Keeping diagnosis and validation honest and separate from delivery is what makes the causal layer trustworthy.
What happens next is a genuinely open decision, not a handoff we're forcing. Many organisations have strong delivery, governance and measurement capability already, and are happy to take validated improvements forward themselves. Others want the same team that ran the diagnostic to carry it through, and our digital product, AI and service model design capability means we can do that without breaking the thread between what the diagnostic found and how it gets built. Either way, the causal layer stays intact. It is the discipline that does not change; only who picks up delivery does.
If you're not sure whether your CX programme is starting in the right place, that's a conversation worth having. And if you've watched CX investment fail to shift the dial, it's definitely a conversation we should have.
Get in touch