Your customers keep telling you they already told you something. They filled out the form. They explained the issue to your support team. They gave you their preferences. Then they have to start over when they reach the next department.
You know this is broken, so does everyone else. But the question most organizations get wrong is what exactly is broken.
At a recent MarTech Conference panel on building context-aware customer experiences, moderated by Annette Franz, CEO of CX Journey Inc., I joined Shiv Gupta, principal at Quantum Sight and Haley Trost, director of product marketing at Braze. Every panelist and the majority of the audience agreed on the culprit. It wasn’t the technology. It wasn’t the data. It was the operating model.
That should make you uncomfortable. Because fixing an operating model is harder than buying a new platform.
The activation gap nobody wants to own
Here’s the pattern I see constantly: an organization invests in a sophisticated martech stack, runs a solid implementation, maybe even gets the integrations working. Then nothing changes for the customer. The data exists. The tools work. But nobody has figured out how to activate any of it when it matters.
The 2025 State of Martech report found that 61% of organizations still struggle to activate their data effectively, even after investing in CDPs and orchestration tools. That’s an execution gap that lives inside your operating model.
Trost nailed this when she used the word “activation.” It’s not about whether you have the data. It’s about whether your teams can use it in real time, across every channel, when the customer actually needs you to know who they are. Without an operating model that defines how you turn strategies into action, your super-sophisticated DXP is just a super-expensive CMS.
Your customers search everywhere. Make sure your brand shows up.
The SEO toolkit you know, plus the AI visibility data you need.
Start Free Trial
Get started with
Stop chasing the 360-degree unicorn
I’ll be the contrarian here, just like I was on the panel. The 360-degree view of the customer is a myth. I’ve been doing this for over three decades and I’ve yet to see an organization pull it off.
Your customers are constantly changing. Preferences shift daily. Needs evolve based on context you can’t capture. The idea that you can assemble a comprehensive profile at any given moment just doesn’t hold up in practice.
What does work is picking one authoritative system and honoring the most recent signal. You don’t need perfect data. You need actionable data. Get to 80% and you’ll be way ahead of competitors still arguing about which system should be the source of truth.
Shiv reframed the 360-degree view as a question of purpose: what do you actually need it to accomplish? If you focus on solving 80%-90% of the problems in a customer experience, you don’t need every piece of data you’ve ever collected. You need the right data, clean enough to act on.
A panel at the September MarTech Conference on data activation surfaced the same challenge. When polled, attendees identified data accessibility as their biggest activation challenge, with integration as the weakest link in their stacks. The pattern keeps repeating because organizations treat customer data as a static storage problem when it should be a dynamic signal they respond to in real time.
Nobody owns the customer (and everyone knows it)
Here’s a question I posed on the panel: Can you tell me who owns the customer in your organization? I bet you can’t. It’s not a reflection on you personally. It’s how most businesses operate.
Marketing is measured by lead volume. Sales is measured by revenue. Support is measured by resolution time. None of those metrics requires those teams to work together. Worse, they’re compensated on completely different measures.
The fix is simpler than most people make it out to be. Give your teams a KPI they can’t hit on their own. Maybe it’s whether the customer came back and bought again. Maybe it’s lifetime value, which the entire panel agreed was a strong candidate. Your departments will never voluntarily break down silos when their bonuses depend on maintaining them.
The most common CX mistake brands make is starting with tools instead of customers. When departmental metrics outrank customer outcomes, even the best technology can’t compensate.
A CDP won’t save you (and that’s OK)
I’m not a CDP hater. But the CDP industry has done a remarkable job making you believe their product is a prerequisite to connected customer experiences. It’s not.
You probably already have a platform that can serve as a system of record. It might not be perfect. It doesn’t need to be. Before you buy another tool, ask yourself two questions: have you maximized the investments you’ve already made? And who in your organization owns data quality?
If you can’t answer that second question, a CDP won’t fix your problems. It will just give you a more expensive place to store the same messy data you already have.
Shiv offered a reframe worth considering. Sometimes the value of a CDP isn’t the technology but the discipline it forces. When you sit down and say, “This data set is different from my marketing database, different from my CRM and it’s designed to execute something specific,” you’re doing the strategic work that actually matters. The tool is secondary to the thinking.
What works
You know your connected experience strategy is working when customers stop repeating themselves. That’s it. When your service rep knows what a customer read, clicked or did before they called, the handoffs are doing their job.
Customers rarely applaud a seamless experience. They just stop complaining. The absence of friction is your real success signal. Fewer support tickets. Better retention. Silence that speaks louder than any NPS score.
Getting there requires the operating model work nobody wants to do. It means defining who owns the customer, who owns data quality and how teams are compensated. It means choosing a system of record rather than chasing architectural perfection. It means accepting that 80% of a 360-degree view, activated in real time, beats a complete profile gathering dust in a database.
Your technology probably works fine. Your operating model doesn’t. Fix the model first. The stack will follow.
