Reflecting on a year defined by fluctuations in DTC healthcare, AI, and fragmented viewing environments, one key takeaway emerges — you’re only as good as your data. While many went deeper on AI, the importance of usable, clean, privacy-centric, and relevant data implemented responsibly and delivered across the ecosystem came to the fore. The foundation for health innovation and AI depend on comprehensive and interoperable data.
To that end, instead of focusing on a single partnership or breakthrough, our focus is on expanding and harmonizing healthcare data so that visionaries across the industry can build the next generation of healthcare solutions.
Nearly every major initiative we took on — across life sciences, healthcare advertising, payer/provider, and technology — was rooted in the same idea: no organization builds the future of healthcare alone, and no dataset delivers its full value in isolation.
Healthcare data is fragmented by design. Clinical notes explain why something happened, while lab data adds specificity, pricing and coverage shape access, and advertising exposure influences awareness. Each dataset answers a different question, and none of them live in the same system.
For instance, while claims are foundational to how we understand healthcare, they’re only one part of a much larger picture.
To solve this, we’ve focused on what we do best: acquiring data at scale, harmonizing it with clinical and analytical rigor, and making it usable inside the platforms where real decisions are made. Adding to that, our clients want to understand what’s likely to happen next — when a health journey begins, how it unfolds, and where intervention actually matters. That requires earlier signals, more context, and data that can be connected across systems that were never designed to work together.
Over the past two years, the scale of our data has expanded dramatically, multiplying in volume. Our datasets are rooted in claims covering over 330 million de-identified patient lives across billions of healthcare records. That growth and our commitment to creating one of the most robust datasets on the market have allowed us to service the top pharma companies and power hundreds of diverse partners.
In healthcare advertising and life sciences, that means working closely with partners across the ad tech ecosystem to enable privacy-forward activation and measurement at scale. Our role has been to leverage target segment data and make outcomes-based measurement possible, generating meaningful insights.
On the payer/provider side, partnerships have allowed us to combine claims with pricing transparency, remittance, and hospital data. The result isn’t just more information — it’s a clearer view of cost, access, quality, and utilization across healthcare systems and networks, delivered through tools designed to answer specific operational and strategic questions.
Across all of these efforts, a consistent pattern has emerged: the value of our data increases exponentially when it’s integrated with complementary datasets and delivered through the platforms people already use.
Internally, we often describe our data strategy using three words: quantity, quality, and liquidity. We feel good about the first two. The scale is there, and the data is clean, harmonized, and analytically sound. Where we spent much of 2025 — and where we’ll continue to invest — is liquidity.
Liquidity is about how easily data can move into the hands of smart people doing interesting things with it. We’ll never be the best at everything, but if we can be the best at data — and at connecting that data to other systems — we benefit everyone involved. Insights don’t come from data sitting in isolation. They come from data in motion, flowing securely between organizations, platforms, and analytical models.
That focus on liquidity naturally intersects with AI, a topic that was impossible to ignore in 2025.
Given our data foundation, we know that models get smarter, but they aren’t useful without the right data informing them. The performance of any AI system is fundamentally tied to the data it’s trained on and connected to. In healthcare, that challenge is amplified as signals live in separate systems but need to be combined for the greatest impact. Our acquisition of KAID Health reflects how we think about AI in this context. By combining large-scale structured claims data with AI-driven extraction from unstructured clinical notes, we’re beginning to bridge the gap between events and experience — between diagnosis codes and how patients actually progress through care.
Looking ahead to 2026, we’ll continue expanding our datasets, investing in infrastructure that improves access and auditability, and deepening partnerships across the healthcare ecosystem.
Our goal is to make healthcare data more usable, more connected, and more impactful — so that whatever gets built on top of it, whether it’s analytics, AI, or something we haven’t imagined yet, actually works in the real world.
Because when data works, everything works better — decisions, investments, and most importantly, patient outcomes.
VP, Strategic Partnerships
Ted Sweetser is a seasoned professional in the advertising and ad tech market and currently serves as the VP, Strategic Advertising Partnerships at PurpleLab. In PurpleLab, Ted has played a crucial role in building and driving the advertising vertical strategy to ensure the company’s success within the sales and customer support realm. Now leading a growing team of sales executives concentrating on delivering value to advertising and ad tech, Ted has firmly placed PurpleLab into a market-leading position with innovation and continued market expansion.
VP Data Ecosystem and AI Partnerships
Matt Ryklin joined PurpleLab five years ago after growing an early AI startup through acquisition. At PurpleLab, Matt built out and led the Solutions Engineering team, combining data savvy with sales strategy to ensure PurpleLab clients were always set up for success. Now, Matt is focused on building partnerships that make PurpleLab an industry leader in connectivity and interoperability.