Across financial services and banking, AI in customer experience is no longer an experiment — it's an operating assumption. Self-service bots handle routine balance inquiries, agent-assist tools surface next best actions in real time, and quality management is increasingly automated.
The technology is in place. The confidence is high. The ROI? That's where things get interesting.
Consider a bank that proudly reports a 40% deflection rate: its chatbot is handling thousands of inquiries without a live associate. What it can't tell you is whether any of those customers actually resolved their issue, or if they simply gave up and called back. Or a fintech that wants to A/B test a new AI-powered onboarding flow but can't, because compliance restricts the customer data the experiment would need.
The AI is running but the outcomes are invisible. That's the measurement gap, and it's where the ROI story breaks down for many financial services companies.
A new research study conducted by TTEC Digital and CX Dive, "The Great CX Reset," shows the industry has crossed the threshold from adopting AI to operationalizing it. Financial services leaders are confident, their tech stacks are relatively unified, and they can point to where AI is being used across the customer journey.
But for all that maturity, the returns remain hard to measure. And the very things that make banking unique (privacy, security, compliance) are also the things holding the data foundation back.
The result is a sector that's ahead on deployment but stuck on value. This is the great CX reset: the moment when financial services leaders stop asking "Are we using AI?" and start asking "Is it paying off — and how do we know?"
Confidence is high, and earned
If there's one thing the data makes clear, it's that the financial services industry isn't suffering from an AI confidence problem. 100% of industry respondents say they are very or somewhat confident in their organization's ability to improve CX with AI.
And that’s backed by real deployment. When asked how they're currently using AI in CX, the top answers were: customer self-service (93%), quality management (88%), agent assist(88%), and personalization (65%).
AI is already touching the full customer journey, from the first self-service interaction through to quality oversight of live conversations. And financial services leaders believe they can pinpoint it: 91% are somewhat or very confident their organization can clearly account for where AI is being used in the customer journey.
That visibility matters. Across the broader report, which surveyed 150 CX, contact center, and IT leaders across various industries, teams with real visibility into where AI operates are 52% more likely to believe AI will deliver on its promise.
The tech stack is more unified than you'd think
One of the quiet strengths of the financial services AI story is the underlying architecture. When asked to describe their tech stack, 83% of industry respondents said they are either mostly unified on a small number of integrated platforms or using a core set of integrated platforms plus some point solutions.
For a sector historically weighed down by legacy systems and siloed data, that's a meaningful signal. Integration is no longer the primary obstacle it once was, which means the remaining barriers to value aren't about infrastructure; they're about proof, governance, and the data itself.
The barrier isn't adoption. It's measurement.
When asked to name the biggest barrier to getting more value from AI in CX, the greatest share of financial services respondents (50%) pointed to difficulty measuring ROI.
This is the central tension of the financial services AI story: deployment is mature, confidence is universal, the stack is integrated, and yet the industry can't confidently connect AI investment to financial outcomes. It's not that AI isn't working; it's that the measurement framework hasn't caught up to the deployment footprint.
In a sector defined by rigor, flying blind on AI ROI is unsustainable. The reset will be led by the institutions that build the measurement discipline to match their deployment ambition.
Privacy, security, and compliance: the data challenge that defines the sector
When asked which data-related challenges most limit their organization's ability to realize value from AI in CX, 53% of financial services respondents pointed to privacy, security, or compliance restrictions limiting how they can use data.
This is a cost of operating in one of the most heavily regulated industries in the world. Financial services institutions should be cautious with customer data. But the consequence is real: the very guardrails that protect customers also constrain the data foundation that AI needs to personalize, predict, and improve.
The institutions that win the next phase won't be the ones that loosen the guardrails. They'll be the ones that modernize the data foundation within them, building compliant and governed pipelines that make data usable without making it exposed.
Governance: in place, but not always in practice
The data tells a familiar story here. When asked about AI governance, 60% of financial services respondents said they have some governance policies but they are sometimes applied inconsistently.
For a regulated industry, "sometimes" is a risky word. Inconsistent governance creates blind spots and blind spots create risk. The institutions that close this gap will treat AI governance not as a compliance checkbox but as an operating discipline, applied the same way in every queue, every channel, every geography.
Where financial services leaders are pointing next
When asked where their 2027 priorities lies, the most common answers from industry leaders were modernizing the customer data foundation (50%) and using AI to improve efficiency and productivity (43%).
Financial services leaders know the deployment side is largely solved. The next frontier is the foundation underneath it, the data layer that determines whether AI can actually deliver measurable value. Efficiency and productivity gains will follow, but only for the institutions that fix the foundation first.
The financial institutions that win won’t be the ones with the most AI tools; they’ll be the ones that can prove whether any of them worked.
The Great CX Reset lays out the full picture, with breakdowns by leadership role and industry, a clear-eyed look at where AI initiatives are stalling, and a closer look at what the strongest teams are doing differently. Download it here.
