Why is CX investment in AI outpacing ROI?

By Katie Judd, General Manager at CCW Europe

Any debate over whether or not artificial intelligence has or will have a place in CX has been put to bed. AI adoption is now a consistent focus of boardroom-level discussion, as organisations embrace its potential to transform operations, drive growth, and retain customer loyalty. 

New data from CCW Europe’s community has found that 66% of CX leaders now say AI adoption is a matter of high strategic imperative. The race to adopt is well underway, and IT budgets have shifted accordingly; 65% of CX leaders said they expect AI to receive the lion’s share of new CX investment over the next one-to-two years. Organisations are spending heavily on new AI tools, launching pilot programmes, and working to keep up with competitors’ announcements of AI success. 

So, where are the returns? 

Despite significant spending, over half (57%) of CX leaders say that AI has had limited or even no impact on their operations. In fact, just one-in-ten CX leaders felt as though AI was making a significant difference.  

There is a growing gulf between AI deployment and the commercial impact in CX. Closing it means understanding why investment isn’t consistently translating into outcomes, and then building the necessary foundations that let AI break out of the pilot stage.

The AI ROI gap and why it exists

In spite of strengthening executive commitment, 62% of organisations are still in the earliest stages of the AI adoption curve. At the other end of the spectrum, just 22% have meaningfully scaled their adoption, with 16% embedding AI within key workflows and 6% using it as a genuine source of competitive differentiation.

Our data suggests three persistent barriers to successfully integrating and scaling AI in CX. 

First, measuring success (or failure) isn’t entirely straightforward, with 21% of organisations struggling to link AI-related efficiency gains to tangible financial outcomes, and 18% finding it difficult to demonstrate ROI beyond an individual use case. Time saved, reduced human input and faster processes can all represent meaningful gains, but putting a price tag on those improvements can be frustrating in a large, complex organisation.

Secondly, objectives for AI implementations are often too broad. The sheer scope of AI’s potential makes specificity more important, not less. Organisations need to define what they are trying to change, establish the metrics that will demonstrate progress, and agree in advance what success will look like. Without that degree of discipline, AI programmes can accumulate use cases while remaining disconnected from strategic priorities.

Lastly, organisations consistently struggle when transitioning from isolated pilots to AI adoption at scale. AI works very well in the pilot phase, when inputs are more easily controlled and outputs are more carefully monitored. Pilots do a good job of demonstrating that a technology works in theory. When the rubber meets the road at scale, however, those theories have a habit of falling apart.

The organisations seeing real success right now are the ones elevating their AI initiatives from a lone CX initiative to an enterprise-wide endeavour. Their AI implementations are built with scaling in mind from day one, which reduces or eliminates entirely the need for retrofitting once a pilot has found its feet.

Closing the AI ROI gap

The first, and perhaps most important, requirement is enterprise-wide commitment.

The organisations making the fastest progress are those where executive sponsorship develops into real C-suite ownership. This allows technology teams to provide infrastructure and technical expertise, data teams to maintain pipelines, models, and data quality, legal and compliance teams to manage regulatory exposure and procurement to manage the commercial foundations.

Leading organisations are increasingly establishing dedicated AI and CX leadership capabilities to bridge these disciplines. Often positioned close to the C-suite, these teams combine deep CX knowledge with practical expertise in technology and AI. They can set enterprise standards while maintaining accountability with the teams closest to the customer.

Instead of allowing AI pilots to develop in the rarefied seclusion of an IT department, this approach allows strategic decisions to sit closer to the business. From day one, someone is asking which problems to solve, which use cases deserve priority, how customer journeys should change, and how to measure value. 

The second requirement is greater precision concerning expectations for AI’s achievements. Leaders need to narrow the field, define the business problem, and establish the desired customer and commercial outcomes. Set a baseline, and agree which metrics will be used to demonstrate value.

The issue is no longer whether AI can perform a particular task. The question is whether your organisation is designed to capture the resultant value.

Where do we go from here?

AI has huge potential to transform CX, but the journey is far from over. No organisations in the current data reported enterprise-level impact from AI in CX at a material scope across revenue growth, retention, and margins. Less than one-in-five (18%) report no quantifiable impact to date, while a further 39% describe gains that are real but remain isolated. Only 11% report functional successes with measurable commercial outcomes.

For CX leaders, the next phase of the AI journey is less about proving what the technology can do. It is now about ensuring their organisation is capable of turning those capabilities into measurable outcomes, repeatedly and at scale.