Life sciences companies are making artificial intelligence (AI) a priority—not only in drug development but also in commercialization. A recent Accenture survey found that 90% of industry executives recognize AI as important in driving commercial innovation such as hyper-personalized experiences. We are on the cusp of extraordinary AI breakthroughs, but there’s a missing link—the right data foundation.
AI needs data to build intelligence and for that data to be useful, it must be organized in a standard way. For decades, the industry has relied on building and maintaining its own custom-built data warehouses because of a lack of high-quality, packaged commercial data warehouse solutions designed for the life sciences industry. These custom solutions are not easily adaptable to changing data structures and sources so companies cannot fully leverage the rich data available in real time. Further, this delays insights from reaching the business and leaves organizations ill-equipped to support AI.
A next-generation commercial data warehouse solution solves the challenge of creating and maintaining a data foundation for analytics and future AI. Designed specifically for the life sciences industry, this cloud-based technology provides a standard model that brings together all of the industry’s most important data sources and organizes them in a way that makes it more efficient to analyze. With this data foundation, life sciences companies are equipped to apply AI to their data, identify meaningful patterns, and draw valuable learnings.
Recently, three industry leaders discussed how this new model will address today’s data challenges and elevate their commercial operations with the right foundation for AI. The participants in this discussion were:
Jason Magyar, IT Director at Karyopharm Therapeutics, a pharmaceutical company focused on the discovery, development, and commercialization of medicines with the goal of improving the lives of patients with cancer. The company’s primary focus is on developing novel drugs to help treat patients with certain blood cancers or solid tumor malignancies.
Salvatore Paolozza, Director of Sales Operations at Antares Pharma, a specialty pharmaceutical company that combines pharmaceutical and medical device expertise to develop innovative products that address needs in underserved therapeutic areas. Two of their products include Xyosted for testosterone replacement and Otrexup for rheumatoid arthritis.
John Chinnici, General Manager for Data Management at Veeva Systems, a leader in cloud-based software for the global life sciences industry that recently introduced Veeva Nitro, a cloud-based data warehouse solution.
What are the industry’s expectations for commercial data?
Jason Magyar: “The expectations for using business intelligence have dramatically increased in the past few years. In the past when launching a drug, an organization had limited data and business intelligence requirements. You reported on sales using a lean database of information. Now, with the abundance of data, day one reporting requirements are dramatically more complex and voluminous. As that trend continues, it’s important that startups establish a business intelligence capability in the fastest and most efficient way—and that is what is so important about a pre-packaged data warehouse.”
Salvatore Paolozza: “Every pharmaceutical company wants to get accurate data into the hands of their sales reps as soon as possible so it remains actionable. This requires the right technology to organize data from many different sources and serve it to reps as soon as possible.”
John Chinnici: “Organizations want to generate insights across their business quickly. They want to fully leverage all of their data investments and all of their sources of data in real time to perform advanced analytics for intelligent engagement with customers. Unfortunately, they haven’t had the tools to do it.”
What are industry’s biggest data management challenges?
Magyar: “As an oncology-focused pharmaceutical company in the midst of launching our first commercial product, we anticipated a flood of data from the get-go and needed a place to store it in an organized fashion. Our biggest challenge was building a business intelligence foundation for our future.”
Paolozza: “With a lot of data coming from many different companies, we were having issues bringing it all together and then feeding it to our reps. At Antares, we were delivering data through a number of different ways such as Excel sheets and our CRM system. Reps had to go to three or four locations to retrieve it. It was messy, time-consuming, and late.”
Chinnici: “Historically, the biggest has been that companies spent a lot of time and resources building their own customer data warehouses. These data warehouses don’t evolve as quickly as the business, making them increasingly difficult-to-use and the data outdated—sometimes from day one. Further, these custom solutions are not flexible enough to manage the changing data needs with both unstructured and structured data sets that we have today.”
What are the key considerations for adopting a new data solution?
Magyar: “In building our data warehouse capabilities, we didn’t want to write a myriad of complicated specs for every data element or contract a vendor to map out every field in the database. We also didn’t want to buy data storage or source integration tools. At Karyopharm, we wanted a plug-and-play solution that would put us on the fast track without a huge investment.”
Paolozza: “We wanted technology that would allow us to aggregate all of our data in one location, analyze it, and swiftly serve that up to our reps. The ideal data management solution runs on its own and has the least number of touchpoints. I wanted one system where my reps could look at information, take action on it, and get feedback.”
Chinnici: “Organizations need a powerful platform built to make it easy to generate insights that drive business strategy. Deriving insights from a disjointed or disconnected data platform is impossible. The challenge is organizing data on the right platform to do AI and intelligent engagement.”
How do previous generation data warehouses compare with current data needs?
Magyar: “We didn’t consider the traditional route as implementation would take too long and cost too much compared to a pre-packaged data warehouse solution. I’ve seen custom data warehouse projects cost organizations upwards of $1 million—and fail. We spent a fraction of that and delivered a robust data warehouse from the get-go.”
“Developing a traditional data warehouse is like building a house that requires having all the material and sources, but a modern cloud-based system already has all the building materials and blueprint. Right out the gate with our next-gen commercial data warehouse, we could bring in all our CRM data so we were able to deploy customer insights to the field in less than three months. And, every new data source is automatically connected. In the past, just integrating external data sources would take months.”
Chinnici: “Data is complex and difficult to organize. The first data warehouse solutions were built from scratch for each company without rich analytics capabilities. Analytics-ready data warehouse solutions can bring together data sources so users can implement the business intelligence tools of their choice.”
Paolozza: “It wasn’t good. We had a custom data warehouse that was expensive to maintain when making any changes, whether bringing in a new data source, retargeting parameters, or modifying information sent to reps. You don’t know what you will need six months down the line. By designing the data warehouse to meet our reporting needs, we also limited its capabilities to just what we knew.”
How does a data warehouse support advanced analytic functions like AI?
Magyar: “Organizations must structure data to deploy the true vision of AI. Until then, its use is confined to specific cases. But, the potential benefits of AI to sales productivity are tremendous. As a startup, we buy syndicated data on the customer universe to identify target audiences. With our focus on oncology, we also buy claims data on patients with cancer to understand different therapies during the first line, second line, etc. With this information factored in the data warehouse, AI can process billions of records to segment target prescribers. It allows us to get a jumpstart on commercialization to craft marketing messages and sales tactics that address different types of biases among prescribers.”
Paolozza: “The power of AI is identifying and profiling customers, then serving them with information to adopt your product. With intelligence incorporated into reporting, we can profile the right physicians who can write our products and treat their patients. While we aren’t there yet, we are starting to get some benefits where we can bring in daily data and send intelligent solutions to our reps. Of our 83 reps, 10 took immediate advantage of the solution and rose to the top of the pack within a couple of weeks. By using a pre-packaged data warehouse solution, we get all the benefits of enhancements such as AI that would not be available in a custom data warehouse.”
Chinnici: “The inability of custom-built data warehouses to quickly adapt to changing data structures and sources leaves organizations ill-prepared to support AI. A next-generation commercial data warehouse forms the foundation for intelligent customer engagement by standardizing many discrete data sources, including CRM, claims, prescription, and more to fully leverage the power of advanced analytics. Users can implement BI and AI tools as well as tailor data visualization to field teams to immediately generate insights and drive informed action.”