SPEAKERS:
· Amy Campion, Senior Vice President of Business Development and Strategy, Timely by DrFirst
· Mayank Misra, VP Commercial Strategy, Analytics and Operations, Soleno Therapeutics
· Venera Jordan, Global Capability Partner, Omnichannel and Digital, AstraZeneca
· Gaurav Jaggi, Director, Strategic Insights & Analytics, Bayer
KEY TAKEAWAYS:
• Thirty percent of referring physician data vanishes in complex therapeutic referral chains
• Big pharma prioritizes longitudinal patient journeys through privacy-preserving data integration
• Consumer data proves essential for reaching caregivers invisible in clinical datasets
• AI acceleration demands data foundation investments before model sophistication
• Direct-to-consumer models reshape pharma from B2B to B2C with transparency expectations
Prescriptions disappear into "the ether" after doctors write them, leaving patients confused about costs, access, and next steps. It's a moment of maximum vulnerability—and maximum opportunity. Amy Campion opened the Pharma Customer Engagement USA 2025 panel by highlighting this critical touchpoint. "We deliver a copy of that prescription to the patient as well as to the pharmacy. It allows folks to understand what they will pay," she explained at the Philadelphia conference, describing how Timely by DrFirst bridges the information gap. The panel's mission: help pharma companies capture customers at these fleeting first moments before they're lost to confusion, cost barriers, or competitor alternatives.
When 30% of Your Customer Journey Disappears
Pharma companies invest millions in data infrastructure, yet critical gaps persist in understanding how patients navigate treatment pathways. Mayank Misra from Soleno Therapeutics revealed a sobering example:
"We had about 30% of our referring physician data missing. And the reason is because when the patient referral happens very often the referring physicians don't end up writing their name."
In radiopharmaceutical therapy, this creates a "black hole" between medical oncologists or urologists and the nuclear medicine doctors who ultimately treat patients.
Even county-level mapping produces 1:20 ratios between referring and treating physicians—too broad for actionable targeting. Gaurav Jaggi from Bayer described solving similar universe coverage problems through data layering: "We combined other data sets like PDS data and others together up to 75, 80% accessibility, which really helps in targeting more effectively for the drug." His team started with SQL data capturing only 50% of relevant HCPs in certain disease areas.
Visibility gaps extend beyond physician identification. Jaggi explained how his team uncovered therapy discontinuation drivers: "We were able to only find out why patients were not adhering to the drug for as long as we would expect by again layering on data sets like EMR which really gave us insights into comparisons between first-line, second-line settings." Claims data showed adherence patterns, but only EMR integration revealed the clinical context explaining patient behavior.
These examples underscore a fundamental truth: single data sources, no matter how comprehensive, cannot illuminate the full customer journey. Field teams operate blind on significant portions of patient acquisition funnels, wasting resources on untargeted outreach while missing high-value referral sources.
Reaching the Invisible Decision-Makers
Traditional pharma data strategies focus on patients, but in rare disease, oncology, and serious conditions, caregivers often make treatment decisions—and they're invisible in claims and EMR data. Amy Campion shared a personal perspective: "What you're saying is it's hard to find me because I'm not your patient. And so how do you make sure that we get that information out to the right people at that right moment?" Her experience as a caregiver for a child with a rare disease illuminated the challenge of targeting individuals who don't appear in healthcare datasets.
Venera Jordan from AstraZeneca emphasized this strategic imperative: "It's simply going to the theme of the panel, getting that first touchpoint right. And this means website data and then matching that, linking it to either CRM or other information so that your awareness campaigns are targeted appropriately." For caregivers, consumer data—behavioral patterns, socioeconomic profiles, website interactions—matters more than clinical data.
This realization has profound implications for marketing funnel effectiveness. Companies investing millions in patient journey mapping may be optimizing for the wrong audience. If caregivers drive 40-60% of treatment decisions in certain therapeutic areas but represent 0% of the data foundation, segmentation models and content strategies are fundamentally misaligned.
Jordan advocated for capturing preference data at consent touchpoints: "There is a future in that first-party data... through that exchange of value, we were talking about it earlier, how would this value exchange work? But through that, if you get it right, you can build a very good database." This approach—asking patients and caregivers about content preferences, channel choices, and data sharing comfort levels—creates databases that, when linked to third-party data, become more powerful than either source alone.
Why "Garbage In, Garbage Out" Still Applies in the AI Era
Despite enthusiasm for AI applications, fundamental data quality principles haven't changed. Mayank Misra cautioned: "AI still requires the right data. It still requires quality data. It doesn't help if they talk about needle in a haystack. It doesn't help you to double the haystack." Companies rushing to implement generative AI for rep training or next-best-action may be building on unstable foundations.
Gaurav Jaggi described the technical architecture required: "You have a semantic layer. We were talking about using a common language in a couple of sessions earlier today. It helps these databases process in common terms, the same definition for new patient starts, persistence and so on." Without this foundational work—data virtualization connecting databases, metadata structures enabling AI processing, and common definitions across sources—AI models produce inconsistent outputs.
The challenge extends beyond technical infrastructure to organizational alignment. Jaggi emphasized: "Having commonality like common ontology, for example, or common taxonomy between, for example, again between R&D, medical and commercial. Like disease, disease stage, line of therapy biomarker." When different functions define terms differently, AI cannot synthesize cross-functional insights.
Misra's team addressed decision latency by pushing insights directly to field teams: "We are feeding our field with AI-enabled insights into their inboxes... so that the decisions can be made right at the first step." This near real-time approach overcomes the 2-3 month lag in traditional reporting.
From B2B Model to Consumer-Grade Experiences
The pharma industry faces a trust deficit. Venera Jordan noted: "Such a low net promoter score for pharmaceutical industry. So I think really engaging providers and patients is key to building this trust and I think data privacy plays a very big role." Consented engagement that respects customer preferences becomes a competitive differentiator in this environment.
The industry's evolution from business-to-business to business-to-consumer models represents both opportunity and imperative. Mayank Misra observed:
"Pharma was always a B2B model and we always wanted to be B2C and it's kind of becoming, like there is a ray of hope there that hopefully the Lilly and Novos of the world succeed."
Companies like Eli Lilly, Novo Nordisk, and Pfizer are pioneering direct fulfilment with real-time tracking and next-day delivery—matching consumer expectations from retail.
Amy Campion captured patient frustration with traditional models: "I think about pharmacy. Go to pharmacy, okay, I have to wait a week or go all the time. So them pulling their hair out. Why is it not possible today?" Patients accustomed to Amazon-level transparency are increasingly unwilling to tolerate opaque pricing and slow fulfilment.
Venera Jordan articulated the strategic focus for big pharma: "The longitudinal patient journey, creating those journeys is probably where most of big pharma is focusing right now. It's really privacy-preserving linkage between all the data that we need to create that profile." This integration of claims, EMR, lab results, specialty pharmacy, and patient support hub data, done with appropriate consent and privacy protections, enables companies to identify care gaps and optimize engagement at scale.
Building the Foundation for First-Touchpoint Excellence
The panel's insights converge on a central truth: capturing customers at first touchpoints requires data strategies that are simultaneously more integrated and more respectful of individual preferences. The technical challenges—semantic layers, common ontologies, privacy-preserving linkage—are solvable. The organizational challenges, breaking down silos between clinical, medical, and commercial functions, require executive commitment to shared data stewardship.
Three imperatives emerged for pharma leaders. First, audit data blind spots systematically. Whether it's missing referral chain information, caregiver profiles, or indication-level granularity, companies must quantify what they don't know before investing in AI sophistication. The 30% referral data gap at Soleno and the 50% HCP coverage limitation at Bayer represent the kinds of specific, measurable problems that data layering can address.
Second, build first-party data capabilities through value exchange. As direct-to-consumer models create new patient touchpoints, consent-based preference capture becomes strategically critical. Website visits, prescription fulfilment moments, and patient support program enrolment offer opportunities to ask about content interests, channel preferences, and data sharing comfort levels—building databases that complement third-party sources.
Third, prioritize data foundations before scaling AI investments. Near real-time data pipelines, semantic layers enabling natural language queries, and cross-functional ontologies create the infrastructure for AI to deliver "faster, better, cheaper" insights. Without these foundations, companies risk implementing AI that confidently delivers incorrect recommendations based on flawed data.
The stakes are clear: in an era when patients expect consumer-grade experiences and transparency, companies that capture customers at first touchpoints with relevant, timely information will build trust and loyalty. Those that miss these moments will lose patients to competitors who don't.
To get you highlights of Pharma Customer Engagement USA faster, we are using generative AI technology to summarise the transcripts of the sessions. If you have any feedback about the summary, please contact lucy.fisher@thomsonreuters.com.
Discover more on this topic at Pharma USA 2026 (March 17-18, 2026, Philadelphia) - North America's largest cross-functional pharma gathering. Visit the website here.
