Are you struggling to build a reliable commercial data infrastructure to support your first launch?

For pharmaceutical firms, the first drug launch is critical. Without a strategic and technical
blueprint, the launch can be challenging.

What can pharmaceutical manufacturers do to achieve a high degree of success without
risking millions of dollars?

Commercial teams in pharma firms actively seek data-driven answers to build a reliable launch plan. To support this, technology teams must decide on the technology stack, modules to outsource, and steps to be prioritized. They’re not sure where to start and how they can get it right the first time. These high-impact decisions need to account for scalability, data security, and governance, further adding complexity to the launch strategy

Recently, one such pre-commercial oncology brand that was struggling to build a reliable on-cloud commercial data infrastructure, reached out to ProcDNA for help.

ProcDNA experts helped the company dodge the most common mistakes. This saved the company millions of dollars in investment while accelerating their decision-making for building a clear and reliable commercial data infrastructure strategy.

1. Create a detailed roadmap
Our first step was to involve clients’ business teams in setting up roadmaps for short, mid, and long-term implementation. This included listing detailed business questions that they need answers to and defining what success meant for them. Teams often start by creating a stand-alone analytics sandbox to analyze sales and claims data and answer business questions, without realizing that it introduces unnecessary silos into the journey. Analytics teams usually get so used to the sandbox environment that it becomes hard to shift or adapt when the final infrastructure is ready. To help our client avoid this trap, we used our proprietary EDL framework and created the sandbox within the Azure Data Lake environment. Our client also wanted to create a digital-native marketing strategy and wanted to keep all their marketing data in-house. With early directives like these, the company can successfully prepare to build a robust data infrastructure. The team developed the Azure data lake to house all types of marketing content. We then integrated DataBricks into the architecture to enable advanced data science capabilities allowing real-time communication with the MarTech platforms.

2. Adopt modular, best-of-breed platforms
Analytics and advanced data science market is filled with a wide selection of tools to suit different situations. We help our life-sciences clients filter out the noise and choose the right tools for their needs. Our goal is to source best-of-breed tools that are easy to integrate into your overall data infrastructure. Highly modular data architecture allowed our client to use a combination of licensed tools and open-source components that can be plugged in and out as needed. To bypass any complications that may occur in the application layer, most of the integrations with other analytics and visualization tools were done using APIs.

3. Reduce waste with Common Data Layouts (CDLs)
To generate consistent results across distinct functions and to reduce operational overheads, ProcDNA experts created Common Data Layouts (CDLs). The devised Common Data Layout (CDL) allowed the teams to create reusable assets and encouraged the culture of reusability in the organization. Further on, our experts helped the client conceive a list of Critical Business Questions (CBQs). The CDLs were exposed to the application layer using a Role Based Access Control (RBAC) system to control the level of data access for each of the analytics sandbox, advanced analytics, or visualization layers.

4. Design and build to scale
During the lifecycle of a commercial product launch, large volumes of data pour in from multiple channels in a short space of time – product sales, patient claims, call activity, sampling, speaker bureau, branded/unbranded websites, emails, digital campaigns, and others. While it is recommend to design the pipelines in advance with dummy data, companies must also consider scalability to manage these parallel data streams that flush data into the systems in big spurts. To sufficiently manage this situation, data infrastructure teams should be experienced with varied data sources specific to the pharmaceutical industry rather than simply treating it as another IT implementation project. They should test the pipelines with realistic data attributes and volumes. In our experience with multiple life-sciences companies, the absence of real-data testing often leads to failure in meeting critical launch KPIs even weeks after the product launch.

By following the above 4 best practices for building a robust on-cloud commercial data
infrastructure, life-sciences companies can better support new brand launches. With a sophisticated data infrastructure built to scale, data and commercial teams can generate
insights rather than worrying about data operations.