Derek, you’ve built your career at the intersection of data and innovation. What originally inspired you to co-found Amperity, and how has that vision evolved with the rise of AI?
The idea behind Amperity came from seeing how brands were drowning in disconnected, messy customer data. There was a big gap between what data companies collect and what they can do with that data. We wanted to close that gap and make data something brands could confidently act on.
With AI, the need for unified, usable data has only grown. AI sets a higher bar for speed, accuracy, and personalization—but it also amplifies the risks when the data is incomplete or inconsistent. That’s why we’re focused on helping brands use natural language and agentic AI to make data more accessible and trustworthy. The goal is to keep data as the solid foundation that AI depends on.
Many enterprises are racing to adopt GenAI, yet they struggle with scattered and duplicate customer records. Why is fixing identity resolution the first step to making AI effective?
I’m not the first person to say that AI is only as strong as the data behind it. Take, for example, if you have three different versions of the same customer, AI won’t know which one to trust, and everything built on top of that becomes unreliable. Identity resolution is how you fix that because it creates a single, accurate view of each customer by pulling all the fragmented data together. Once that’s in place, AI can do what it’s meant to do: drive personalization, customer insights, and automation at scale.
When customer profiles are fragmented across CRM, e-commerce, and service platforms, what are the biggest risks organizations face in terms of both wasted spend and customer trust?
When data lives in silos, you lose the full picture of your customer. That can lead to wasted spend, like running overlapping campaigns or missing opportunities with your high value customers because their signals are scattered. It also erodes trust customers have with your brand. Service teams might not know about a recent purchase, marketing might send irrelevant offers, and loyalty programs can miss long-time members. All of that adds up to a disjointed customer experience that chips away at loyalty.
AI often gets blamed when pilots underdeliver, but you’ve argued the real issue is usually poor data quality. How do you explain this gap to leaders setting ambitious AI strategies?
AI isn’t magic. It’s more like an architect—you can have an amazing design, but if the foundation is cracked, the whole structure is unstable. Poor data quality is that cracked foundation. Leaders who want to get real results from AI have to start with solid data fundamentals. Otherwise, no amount of investment in models or tools will fix what’s broken underneath.
Personalization is one of the most talked-about promises of AI. How does linking every interaction to a single, accurate profile change what personalization looks like in practice?
Without clean, unified data, AI is just guessing. It sees pieces of a customer’s journey but misses the complete story. When all data is connected into one profile, AI can understand a person’s history, preferences, and context. That’s what makes personalization actually feel personal – it’s relevant, timely, and consistent across every touchpoint.
There’s an ongoing debate around building versus buying identity resolution systems. What factors should enterprises weigh most carefully when deciding which route to take?
It’s not always a simple “build or buy” question—it’s more like “build with.” The key is to figure out which capabilities are core to your business and which can be enhanced by the right partners. For some brands, that might mean building custom workflows on top of an existing platform. What matters most is scalability, adaptability, and making sure the stack you choose can evolve as your business does.
Privacy regulations are tightening globally. How does consolidating customer data into a unified profile actually reduce compliance risks rather than increase them?
When data is spread across dozens of systems, it’s hard to know where everything lives or to honor customer requests like deleting their data. A unified profile gives brands a single source of truth. That makes it easier to apply policies, track consent, and respond quickly to compliance requirements. It also builds a safer, more transparent environment for AI, since you know exactly where data comes from and how it’s being used.
In your work with large consumer brands, what are some concrete examples of how better data foundations have turned AI from hype into measurable business outcomes?
Take New Look, a major UK fashion retailer. They worked with us to unify billions of customer records and discovered 3.4 million fragmented profiles, including 31% of top customers who were previously unrecognized. Once that data was stitched together and shared back into Databricks, they gained a complete, actionable customer view.
With that foundation, New Look launched more precise campaigns, optimized media spend, and improved both in-store and online personalization. The result: a 24% increase in reach to high-value customers and a 50% boost in return on ad spend.
Alaska Airlines is another great example. They unified data across reservations, loyalty systems, mobile apps, and partners to build their first full Customer 360. That data now powers personalized communications tied to real-time flight updates. They’ve seen a 198% lift in conversions, 61% higher open rates, 22% more clicks, and a 30% reduction in ad spend.
As companies prepare their 2026 budgets, how should leaders balance investments between advanced models and the foundational data infrastructure those models rely on?
The higher you want to go, the stronger your base needs to be. Too many teams chase the newest AI tools while overlooking the data quality and governance underneath. You need to start with identity resolution and clean data. Every dollar spent on advanced models without that groundwork risks producing noise instead of value. Once your foundation is solid, everything you build on top moves faster and performs better.
Looking ahead, what excites you most about the convergence of customer data platforms and AI, and how do you see this shaping enterprise strategies over the next few years?
AI is making customer data more usable than ever. Instead of teams writing complex queries or managing siloed systems, we’re entering a phase where natural language, agentic AI, and real-time orchestration are built on unified profiles. It’s changing the way enterprises operate—from running campaigns reactively to learning and adapting in real time.
Over the next few years, customer data platforms won’t just store or process data, they’ll act as intelligent copilots that help shape strategy and guide business decisions.
Derek Slager
CTO and Co-Founder, Amperity
Derek co-founded Amperity to create a tool that would give marketers and analysts access to accurate, consistent and comprehensive customer data. As CTO, he leads the company’s product, engineering, operations and information security teams to deliver on Amperity’s mission of helping people use data to serve customers. Prior to Amperity, Derek was on the founding team at Appature and held engineering leadership positions at various business and consumer-facing startups, focusing on large-scale distributed systems and security.