How to Build the Ideal CX Master Plan for the AI Era

CX Transformed How Institutions Operate. AI Will Transform Customer Relationships.

Customer experience changed how large institutions thought about service. Before CX became a discipline, transit authorities designed routes around operational efficiency, banks designed products around internal risk models, universities designed portals around administrative structure, and government agencies designed services around regulatory compliance. The customer was downstream of every decision.

CX reversed that — or at least it tried to. Over the past fifteen years, institutions invested in the infrastructure to see the experience from the customer's perspective. NPS programs. Intercept surveys. Journey mapping engagements. Analytics platforms. CX teams built the visibility that leaders needed to understand what customers were actually experiencing, where friction lived, and where the gaps between institutional intent and customer reality were widest.

That visibility mattered. Forrester's CX Index has consistently shown that CX leaders achieve 17% compound annual revenue growth compared to just 3% for laggards. The Qualtrics XM Institute estimates that poor customer experiences put $3.7 trillion at risk annually across organizations worldwide. The economic case for CX has never been in question. The ability to act on it is what held most institutions back.

And that is exactly what AI changes.

Where Traditional CX Strategy Hit a Ceiling

If you have led or worked closely with a CX function inside a large institution, the limitations are familiar — not because the discipline was wrong, but because it operated within constraints that the organization was not structured to remove.

Measurement channels were narrow and expensive. CX teams relied on QR codes, email surveys with declining response rates, costly intercept interviews, and analytics dashboards that showed behavior but not motivation. The data was useful, but it covered a fraction of the customer journey. Entire segments — the ones who never filled out the survey, never scanned the QR code, never showed up in the analytics — were invisible.

CX was a reporting function, not an operating one. Most CX teams could tell leadership what was happening. Few had the mandate, the budget, or the organizational authority to change what was happening. The insights went into decks. The decks went into quarterly reviews. And the frontline experience often stayed exactly where it was — because the teams responsible for delivery were in a different part of the org chart, working from different priorities, measured on different KPIs.

Data was siloed across teams, systems, and governance structures. The contact center had one picture of the customer. The digital team had another. The branch or field office had a third. Marketing had a fourth. Each was accurate within its scope and incomplete in ways that made cross-journey understanding nearly impossible. The Salesforce State of the Connected Customer report found that 85% of customers expect consistent interactions across departments — but 54% say it feels like sales, service, and marketing do not share information. That gap is not a technology problem. It is an organizational one. And it has been the ceiling that most CX programs could not break through.

Jochem van der Veer, CEO of TheyDo, put it well on the Product Impact Podcast — an organization's biggest problems are hiding in plain sight, scattered across customer service calls, survey responses, app store reviews, and sales conversations. The data exists. It has always existed. What was missing was the ability to connect it, synthesize it, and act on it at the speed the customer experience demands.

That is the ceiling. And AI is what breaks through it.

How AI Transforms Customer Experience Strategy

AI does not replace CX. It removes the constraints that kept CX from reaching its potential. The shift is happening across every dimension of how institutions understand and serve their customers.

AI Simplifies Access to Customer Data and Makes It Actionable

The data that CX teams spent weeks assembling from disparate systems — contact center transcripts, survey responses, digital analytics, social sentiment, field observations — can now be synthesized in hours. Large language models can process unstructured customer feedback at scale, identify patterns across thousands of interactions, and surface the signals that matter without requiring a data engineering project to connect the systems first.

This does not eliminate the need for human judgment. It eliminates the months-long lag between "we have the data somewhere" and "we understand what it is telling us." For institutions where the CX team's biggest constraint was time-to-insight, AI compresses the cycle from quarters to weeks.

AI Enables New Forms of Analytics That Move Past Surveys and NPS

Gartner predicted that more than 75% of organizations would move away from NPS as a standalone measure of CX success. The prediction was directionally right even if the timeline was ambitious — because AI is making it possible to measure customer experience through behavioral signals rather than asking customers to self-report.

Sentiment analysis across every support interaction. Intent modeling from digital behavior. Effort scoring derived from actual workflow completion, not from a survey question about how hard something felt. These are not experimental techniques anymore. They are operational capabilities that institutions can deploy today — and they produce a picture of the customer experience that is continuous, comprehensive, and far richer than any survey program could deliver.

AI Creates Massive Incentives to Fix Data Gaps and Technical Debt

Here is a dynamic that CX leaders should welcome: AI's effectiveness depends on the quality and connectivity of institutional data. The data silos and technical debt that blocked CX programs for years are now blocking the AI initiatives that executives are spending millions on. The same fragmented data architecture that prevented your CX team from building a unified customer view is now preventing the AI team from building the models leadership wants.

This means the investment case for fixing data infrastructure — the case CX teams have been making for years — now has executive sponsorship from the AI mandate. The work is the same. The budget authority behind it is different. Smart CX leaders are using this moment to get the foundational data work done that was never prioritized when CX alone was making the case.

AI Is Most Effective When Institutions Break Down Silos

The institutions seeing the strongest results from AI are the ones using it as the catalyst to connect what was previously separated.

Bank of America's Erica has surpassed 3 billion client interactions across nearly 50 million users, averaging 58 million interactions per month — not because the AI is remarkable in isolation, but because it connects account data, transaction patterns, and service capabilities into a single conversational interface that spans what used to be separate banking channels. The silo-breaking is the product.

Klarna's experience illustrates the other side of the equation — and it is equally instructive. When Klarna launched its AI assistant in 2024, it handled two-thirds of all customer service conversations in its first month, equivalent to 700 full-time agents. The efficiency gains were extraordinary. But by 2025, customer satisfaction had dropped 22%. The AI struggled with edge cases, emotionally charged interactions, and complex disputes — the moments where customer experience matters most. Klarna is now rehiring human agents and operating a hybrid model. The lesson is not that AI failed. It is that AI deployed without deep enough understanding of where customers need human judgment — without the CX research that maps those moments — will optimize for efficiency and erode the trust that institutions depend on.

McKinsey's research shows that AI-powered customer engagement can reduce cost-to-serve by 20–30% while simultaneously improving customer outcomes. But the reduction depends on integration — AI layered on top of fragmented systems produces fragmented results.

The lesson for institutional CX leaders is clear: the AI roadmap and the CX roadmap are the same roadmap. The data integration work, the cross-functional alignment, the unified customer view — these are not CX prerequisites that need to be completed before AI can begin. They are the work that AI makes possible, necessary, and fundable at the same time.

Competition Is Accelerating — and Institutions Cannot Afford to Wait

There is an external pressure that makes the internal case even more urgent.

AI is compressing the time it takes for smaller, more agile organizations to build digital experiences that rival those of large institutions. A fintech with a fraction of the headcount can now deliver faster, more personalized service than a major bank. A private education platform can build an applicant experience in weeks that outperforms a university portal built over years. A ride-hailing app can offer real-time, personalized transit information that a transit authority's website cannot match.

When PH1 built the CX master plan for TransLink, the pressure from private ride-sharing services was already forcing leadership to rethink how they engaged riders — not just operationally, but experientially. The competitive threat was not another transit authority. It was a private company that could iterate on the customer experience weekly while the institution planned quarterly. The CX master plan defined how TransLink could meet riders across digital and in-person touchpoints with a unified vision — and every year since, new AI capabilities have created new plateaus of potential value delivery on top of that foundation. The infrastructure constraints that once capped what a transit CX program could accomplish — limited real-time data, fragmented rider feedback, channel-specific service — are becoming the institution's greatest advantages as AI turns those data sources into a connected intelligence layer that no ride-sharing app can replicate at system scale.

When PH1 worked with Simon Fraser University, the research revealed that students were accessing information differently and bypassing institutional websites entirely — meaning the institution had to find fundamentally new ways of engaging prospective applicants rather than optimizing the old ones.

When PH1 mapped the creator ecosystem for Spotify, the work identified where analytics features were being used without actually changing creator decisions — engagement without impact — revealing which investments deserved deeper commitment and which needed to be rethought entirely.

The pattern is the same across sectors. The competitive environment is moving faster than most institutional planning cycles. A CX master plan is how you close the gap between the speed at which expectations are shifting and the speed at which your organization can respond.

Yaddy Arroyo, who spent a decade designing AI systems in financial services, put it directly on the Product Impact Podcast — the organizations that succeed with AI in 2026 will be the ones that get three things right: trust, cost, and orchestration. Trust comes from understanding what customers actually need. Cost discipline comes from investing in the right use cases. And orchestration comes from a plan that connects every team to a shared direction. That is what a CX master plan delivers.

What a CX Master Plan for the AI Era Includes

A CX master plan is not a research report. It is an operational commitment — a shared agreement across leadership, operating teams, and digital teams about where the organization is going, what it will prioritize, and how it will measure progress. In the AI era, that plan needs to do five things.

1. Define the AI North Star Aligned with Your Institutional Mission

Every institution exists to serve a purpose that is larger than efficiency. A transit authority exists to move a region. A bank exists to enable financial confidence. A university exists to educate and advance knowledge. A government agency exists to serve the public.

The CX master plan starts by defining what AI-enhanced customer experience looks like in service of that mission — not as a technology deployment, but as an elevation of what the institution can deliver. The north star is not "deploy AI." It is "use AI to deliver on our mission in ways that were previously impossible."

2. Map the Customer Journey with Evidence, Not Assumptions

As we argued in Customer Journey Mapping Is the Secret to Unlocking AI Value, the foundation of any AI investment is a rigorous understanding of the actual customer journey. For institutional CX, this means engaging a wide, representative audience of customers — across segments, across service channels, across stages of the relationship — and building a current-state journey map grounded in observed behavior, not internal assumptions.

The map reveals where AI should augment existing workflows, where it should create entirely new ones, and where it should not be applied at all. It is the evidence base that protects the plan from the most common failure pattern in institutional AI: investing in technology before understanding what customers actually need.

3. Engage Operating Teams, Digital Teams, and Customers in a Shared Vision

The CX master plan is an opportunity — often the first one an institution has had — to align the teams that design, deliver, and support the customer experience around a single picture of what that experience should become. Operating teams bring the reality of how services are delivered. Digital teams bring the understanding of what technology makes possible. And customers bring the truth about what matters and what does not.

The plan must include all three perspectives. Without operating teams, it will be aspirational but undeliverable. Without digital teams, it will be grounded but uninformed by what AI enables. Without customer evidence, it will be internally coherent but externally irrelevant.

4. Redefine How Value Gets Delivered

This is the dimension most CX plans underestimate. A CX master plan for the AI era is not only about making existing services easier or faster — though that matters. It is about identifying where AI enables entirely new forms of value that the institution could not deliver before.

A transit authority that can provide personalized multimodal journey planning — not just routes, but recommendations that account for weather, accessibility, time constraints, and personal preferences — is delivering value that no schedule PDF ever could. A bank that can proactively identify financial risks and opportunities for individual customers — before the customer calls — is delivering a relationship, not a transaction. A university that can guide a prospective student through program selection with the depth and warmth of a great advisor, at midnight, at scale — is delivering access that was previously reserved for the students who happened to reach the right person at the right time.

These are revenue opportunities, retention opportunities, and trust-building opportunities that emerge from the CX master plan when the plan is ambitious enough to look beyond optimization.

5. Build the Measurement Framework That Tracks What Matters

The plan needs a measurement infrastructure that goes beyond the metrics that limited CX in the first place. Not just NPS. Not just satisfaction scores. But behavioral measures of whether the customer experience actually changed — whether task completion improved, whether effort decreased, whether trust increased, whether the AI-enhanced experience produced outcomes the previous experience could not.

PH1 structures this measurement around the Digital Acceleration Pillars — four shifts that separate institutional CX programs that compound from those that stall:

Value — Are you measuring the impact of the experience on the outcomes customers are trying to achieve, or just measuring whether the interaction happened?

Voice — Are decisions grounded in what customers and frontline teams actually said, or in what leadership assumed?

Velocity — Can your organization act on what the evidence reveals in weeks, or does every insight enter a governance cycle that takes quarters?

Vision — Does every team — operations, digital, communications, leadership — share a single picture of the customer experience and a single definition of success?

CX Master Planning Gives Your Entire Organization a Path Forward

The value of a CX master plan is not just strategic. It is organizational.

For leadership, it provides a defensible investment thesis — evidence-based priorities that connect AI spending to customer outcomes, not vendor promises. For operating teams, it provides clarity about what is changing, why, and how their work connects to the larger direction. For digital teams, it provides a roadmap built on customer evidence that prevents the most expensive failure mode in institutional AI: building the wrong thing well.

And for customers, it provides something most have never experienced from a large institution: a digital experience that feels like it was designed for them.

That is the work PH1 was built to do. Fourteen years of CX research and strategy for institutions — transit authorities, financial services, universities, government, and non-profits — delivered by senior researchers and strategists who work across all operating teams to build the shared vision, collect the evidence from across your core audience groups and customer stages, and deliver a development-ready CX master plan your entire organization can align behind. In weeks, not quarters.

Book a Free Consultation →

Sources

1. Forrester, Customer Experience Leaders Crush Laggards on Revenue Growth.

2. Qualtrics XM Institute, Bad Customer Service Threatens $3.7 Trillion Annually, February 2024.

3. Salesforce, State of the Connected Customer, 6th Edition.

4. Bank of America, A Decade of AI Innovation: Erica Surpasses 3 Billion Interactions, August 2025.

5. Klarna, AI Assistant Handles Two-Thirds of Customer Service Chats in Its First Month, February 2024.

6. CX Dive, Klarna Reinvests in Human Talent After AI Chatbot Customer Service Backlash, 2025.

7. McKinsey & Company, The Next Frontier of Customer Engagement: AI-Enabled Customer Service.

8. Gartner, Gartner Predicts More Than 75% of Organizations Will Abandon NPS, May 2021.

9. Product Impact Podcast, Episode 39: The Intelligence Layer That Unlocks Your Business' Biggest Problems (Jochem van der Veer, TheyDo).

10. Product Impact Podcast, Episode 50: Designing AI for 2026 — Trust, Cost, Orchestration (Yaddy Arroyo).