The next age of CX: Maintaining customer facing systems

The marketing landscape is shifting fast, and delivering exceptional customer experiences isn’t optional anymore – it’s survival. During CMA Marketing Week in May, I had the opportunity to sit down with Stephanie Wu, Senior Manager, Audience Data Strategy and Signal Loss Mitigation at BMO, and Darcy Kelley, Managing Director, Data Solutions and Marketing Technology at Omnicom Media Group, to dig into how marketing teams are keeping the voices of their customers top of mind while balancing requirements of legacy customer facing systems with emerging technology.

Both panelists deal with massive amounts of customer data daily. Stephanie manages millions of authenticated data points for BMO's B2C and B2B customers, focusing on personalized experiences across digital channels – from website interactions to email communications and direct mail. As she explained, "The problem that we're trying to solve is more or less in regard to the major headwinds that every marketer is facing today, like evolving cookie policies, user-choice privacy frameworks being implemented by major browsers, and heightened privacy protections from the major technology vendors like Apple, Google and Microsoft."

Meanwhile, Darcy oversees Omnicom's end-to-end orchestration platform, powered by over 20,000 consumer data points and thousands of media response curves. His focus is clear: "Quality of data is super important on the back end. And really, all of the front-end capabilities that are AI powered and enabled all depend on having excellent high-quality data that's properly governed, taxonomized and then provisioned."

The common thread? Change is happening whether we're ready or not, and it's forcing marketers to completely rethink their approach to customer data and personalization.

Making change work in complex organizations

When it comes to implementing new systems, both Stephanie and Darcy emphasized that change management is as much about politics as it is about technology. In my own higher education environment, we deal with decentralized funding, different faculties talking to the same audiences with separate data silos, and a lot of internal coordination just to get everyone aligned.

Darcy's approach is refreshingly practical: test with customer design personas, run alpha and beta phases, and get customer feedback before rolling out broadly. As he put it, "You can spend all day telling people what they should do. And it has the impact of basically communicating that they're not doing their job properly." Instead, he advocates for letting the customer voice drive change and making sure team members are onboarded with that focus in mind.

Stephanie echoed this with BMO's test-and-learn framework. They build audiences, align them with marketing messages, and measure what resonates through dynamic measurement frameworks. Her philosophy is simple: "We want to get a better understanding of what customers really want versus just trying to guesstimate. We also want to deliver more to the ones that are better suited to receive these ideal experiences." But she also highlighted a critical challenge many of us face: getting executive buy-in for future-proofing investments while balancing business-as-usual operations.

The legacy versus innovation balance

One of the toughest questions we tackled in our discussion was how to balance legacy systems that still deliver value with the need to invest in emerging technologies. Stephanie's approach at BMO is pragmatic. They evaluate based on current customer use cases first, then future readiness. As she explained, "Sometimes legacy systems are still driving value. They're still delivering results. Those systems need to be there just to make sure we can migrate smoothly from the old to the new. But we also know that a legacy system won't sustain results after some emerging technology changes."

Darcy emphasized that any new vendor or technology must still bring immediate business value to the table. His criteria are straightforward: "They have to bring something that's net new. And if it's not net new to our current capabilities, it has to be better, faster, cheaper than something we already have." At Omnicom, they put potential partners through rigorous audits including information security, data quality evaluation, and reputation assessments. Many vendors that look impressive on LinkedIn don't make it past the due diligence process.

The AI revolution and what it really means

Here's where things get interesting. Both see AI as transformative, particularly agentic AI that can predict the next best actions and deliver personalized experiences at scale. In five years, Stephanie believes "AI readiness is something that would be a critical measure in evaluating the maturity and the greatness of every martech stack."

But they're clear – AI isn't replacing human judgement. Stephanie made a great point: "AI is great if a task can be replicated, if a task is repetitive, and if we want to achieve the goal of escalation, like personalization at scale. But for experiences where the customers are hoping for more uniqueness, more authenticity, and being more human, then it's not something AI could replace."

Darcy painted a picture of the future where customer journeys become seamless across devices and channels. His vision: "The consumer expectation, when they're using a customer-facing system – whether it's a QR code that they see in a connected TV ad or a website or an app – is that they're able to accomplish what they want to on the first attempt." He sees a world where the painful disconnected experiences we know today become fluid interactions through voice, text or even gestures.

Practical advice for getting started

For organizations looking to prepare for an AI-driven future, the advice was surprisingly straightforward:

Start with your data foundation. Clean, governed data is essential because AI is only as good as what powers it. Have cross-functional teams identifying customer needs that aren't being met, then think about how AI can help deliver solutions.

Focus on privacy compliance early. Stephanie stressed that in highly regulated industries, having your privacy and compliance framework ready is critical for scaling AI initiatives. As she put it, "For any organization, or people who work on data that wants to leverage AI at scale, knowing how to deal with the privacy and compliance and risk teams and having your customer use cases ready from a privacy perspective, is critical to enable the future opportunities." This includes ensuring that all underlying data used in AI systems has been obtained through proper consent mechanisms in compliance with applicable privacy laws, including PIPEDA and provincial jurisdictions, where applicable.

Ask better questions. As AI makes information more accessible, the skill that matters most is knowing what to ask and who to ask. Stephanie emphasized: "One of the critical skills that I personally feel is going to make a major difference is knowing how to ask the right questions, and knowing, when to, or whom to ask the right questions to."

Think customer-first, not technology-first. Both panelists emphasized starting with customer pain points and unmet needs, then working backward to find the right technology solutions.

The privacy and trust equation

We dove into how marketers can meet customer expectations without overstepping trust boundaries. Never capture more data than you need, be transparent about what you're collecting and why, ensure you have proper consent, and provide clear value in exchange for customer information.

Darcy's advice: "Never capture more data than you need to. And if you are capturing data and you're getting consent, it's about being very clear and transparent and using human language with your end customer about what you're collecting the information for."

Stephanie highlighted the value exchange: "Data is a currency. Data is important. We all know that. Let the customer know what the value is of providing this data. If the value is your experience becoming more relevant, more personalized, or enjoying more services, make that value clear."

The bottom line

The future of customer experience is more relevant, personalized and privacy conscious. Customers expect brands to honour their data preferences while delivering genuine value in return. Successful marketing teams will be those that balance AI-powered efficiency with authentic human connections.

Stephanie's final thought resonated deeply: "Despite all the changes and innovations with AI, keeping the human touch is still very important." She emphasized that for marketers, "It's very important to be able to ask the right questions, to identify which are the areas that we should enhance with AI, and which are the areas we still need people to provide a more personal experience to the customers."

Darcy summed it up perfectly: "Customer-facing systems are really all about taking consumers, wants and needs and then putting them at the front of any development. And you really need to be very clear about how you intend to solve their wants and needs and hold your teams accountable to the feedback."

Don't wait for perfect conditions. Keep customers at the centre, start building your foundation now, and be ready to adapt as the landscape continues evolving.

As I concluded during the discussion, the basics of customer experience design remain your starting point: understand who your customers are, listen to their voice, and use that feedback to inform your next actions.