The pharmaceutical industry, like many others, is experiencing a profound transformation fueled by rapid advancements in artificial intelligence (AI). From drug discovery and clinical development to supply chain optimization and new ways of interacting with customers, AI goes way beyond chatbots, and it has the potential to revolutionize every aspect of the life science value chain. 

McKinsey, 2024

AI in pharma is not just a buzzword but a tangible force driving innovation and efficiency. Looking across the value chain in life science, we see new AI projects popping up every week. But how do you as an organization succeed with anchoring AI to truly extract value? How and where do you get started? What pitfalls should you be aware of?

Leveraging AI effectively requires more than just advanced algorithms and vast datasets. It demands a strategic approach, a solid organizational foundation, and a disciplined approach to anchoring and operationalization to translate AI capabilities into real-world benefits.

This article is the first of four where we will post one each week for the next four weeks. In this series, “Successfully Implementing AI in Pharma”, we aim to share our perspective and approach on how to succeed with your AI initiatives, drawing on our experiences assisting leading pharmaceutical companies not only in exploring AI as a trendy must-have, but also harnessing its full potential, achieving tangible benefits, and becoming industry frontrunners in the AI landscape.

While this first article explores the most common use cases we’ve encountered with clients while highlighting a key pitfall: the lack of a strategic view from the outset, the second article will provide a perspective on how to select the use cases that shape your company’s AI vision and ambition in alignment with corporate objectives and feasibility. Then, we will dive into how you can establish a technical and organizational foundation for executing your AI ambition in the third article. Finally, we provide our perspective on anchoring your AI solutions through effective organizational change management.

Opportunities: Common AI Use Cases in Pharma

The opportunities for leveraging AI and generative AI technologies in the pharmaceutical industry are vast across the value chain. By leveraging these advanced technologies, pharmaceutical companies can enhance efficiency, reduce costs, and improve patient outcomes, among other benefits. At Intellishore, we have identified four key areas where we have seen the greatest interest and focus on use cases across the value chain from our clients: 1) Content Creation, 2) Insights Extraction, 3) Predictive Analytics, and 4) Digital Agents.

Content Creation

Generative AI offers transformative capabilities in content creation within the pharmaceutical industry. Traditional processes for creating scientific content, educational material, and commercial content are often time-consuming and intensive on manual labor. AI can automate these processes by generating high-quality, data-driven content rapidly and accurately. For instance, in the commercial space, AI is revolutionizing how pharmaceutical companies engage with healthcare professionals (HCPs) by enabling personalization at scale. For many market players, it has become a strategic pillar to enhance the customer experience (CX) through relevant and engaging content tailored for HCP engagement. Generative AI facilitates the rapid creation of vast amounts of content, allowing for highly personalized interactions with smaller, more targeted segments, all in a timely and efficient manner. AI not only accelerates the content creation process, enabling personalization at scale, but it also minimizes human errors, ensuring a higher degree of consistency and compliance with regulatory standards than human capabilities alone can achieve.

Insights Extraction

Insights extraction in pharma involves the examination of vast amounts of unstructured data, such as free-text insights captured in CRM systems from HCP engagement, research articles, and clinical trial reports. With an AI-driven insights extraction tool, this data can be processed and interpreted much more efficiently than human analysts through Natural Language Processing (NLP) algorithms that extract meaningful insights from text and identify trends. A notable use case involves analyzing insights captured in free-text form by a healthcare company’s representatives during their engagements with HCPs. Throughout the customer journey, HCPs interact with various representatives from a pharmaceutical company, including those from Medical Education, Clinical Trials, and Commercial departments. Each of these representatives can capture valuable insights during their engagements with HCPs. However, due to the sheer volume of daily interactions, it can be challenging to capture, analyze, and utilize this unstructured information effectively. By leveraging an AI insights extraction tool, companies can efficiently process the gathered data in free-text form provided by customer-facing company representatives, extracting valuable insights that can be beneficial for different departments within the organization.

Predictive Analytics

AI-powered predictive analytics has the potential to revolutionize various aspects of the pharmaceutical value chain by analyzing historical data to forecast future outcomes and optimize decision-making processes. With an AI-driven predictive analytics tool, vast amounts of data can be processed and interpreted far more efficiently than by human analysts, utilizing machine learning algorithms to identify patterns and predict outcomes. A notable use case involves optimizing the supply chain for pharmaceutical products. Throughout the production and distribution journey, various factors, such as demand fluctuations, production capacity, and logistical challenges, can impact product supply. By leveraging a predictive analytics tool, companies can analyze historical data, real-time market trends, and external variables to forecast demand and adjust product supply chain operations proactively. This not only ensures a more efficient and reliable product supply but also helps in reducing costs and preventing shortages.

Digital Agents

Digital agents, such as AI-powered chatbots and virtual assistants, can enhance various customer-facing and internal processes in the pharmaceutical industry. These tools can process and interpret vast amounts of data more efficiently than human analysts, utilizing advanced algorithms to provide meaningful insights and assistance. A notable use case involves AI-powered virtual assistants for sales representatives. To ensure a customized CX for HCPs, sales representatives require timely, personalized information to optimize their interactions. However, due to the sheer volume of engagements and data, it can be challenging to provide consistent, tailored support. By leveraging an AI-powered virtual assistant, companies can efficiently analyze preferences, prior engagements, and other relevant customer data to guide field representatives in real-time. This ensures a more personalized and effective HCP engagement, enhances the overall CX, and supports the sales force in achieving better outcomes.

In summary, the application of AI and generative AI across content creation, insights extraction, predictive analytics, and digital agents holds great potential for players in the pharmaceutical industry. These technologies can drive innovation, improve efficiency, and enhance the quality of care provided to patients. But despite all the exciting opportunities AI holds for pharmaceutical companies, implementing AI with success poses some challenges leaders must be aware of.

A Common Pitfall: Why Strategy Matters

We frequently observe AI projects emerging sporadically in silos from different organizational units and teams across clients in the pharmaceutical industry. The fairly recent explosion of generative AI, particularly with the public release of ChatGPT, has inspired many to consider how similar technologies could enhance their daily tasks, making them easier, more efficient, and more in-depth. Consequently, we encounter numerous innovative individuals within our clients’ organizations who have launched impressive AI projects, often securing exceptional traction and funding beyond ordinary budgets. While pursuing AI projects isolated within different functional areas can certainly offer benefits, e.g., due to it being a quick way to pilot a new technology, this grassroots approach with independently initiated projects may not contribute to the long-term success of AI in an entire organization.

A significant pitfall we have observed is that organizations often fail to adopt a strategic, holistic view of AI use cases and analyze potential synergies across the organization rather than focusing solely on developing the technical aspects. While it may seem that taking time to develop a comprehensive strategic view slows progress, especially when others appear to be rapidly adopting AI by “just doing it”, making the initial investment to ensure a structured and holistic approach to AI can ultimately save a substantial amount of time and resources by avoiding roadblocks later on. We observe clear benefits when companies employ a holistic approach towards their AI solutions from the outset by connecting the AI strategy to the corporate strategic objectives and planning how it ought to be anchored by the users in the organization. Thinking about the end-user experience as well as how the AI use cases will be used in the business from the get-go rather than focusing only on the technical implementation significantly enhances adoption and the overall success of the project.

When AI projects emerge within organizational silos, they often fail to adopt a strategic and holistic organizational approach. This often leads to solutions that are narrowly focused on a specific group of internal users or an excessive number of similar projects, each addressing the same business use cases in isolation. By adopting a strategic and end-user perspective towards AI solutions, pharma companies can broaden their scope and identify additional potential users, and identify similar use cases occurring across the organization in the discovery and design phases.

A concrete example we have observed is the development of an AI tool initially designed to analyze free text insights from engagements with HCPs, key opinion leaders, and trial investigators to extract qualitative insights across various therapeutic areas. It emerged from use cases identified in one organizational unit where representatives engage with HCPs frequently and gain interesting insights from real-world settings, for example, patient treatment response and patient preferences and behaviors. When it was launched, it became evident that other organizational units had a large use case overlap for their HCP-facing roles, just based on other types of insights they would gain from HCP engagements. It would have been beneficial for the organization to take a strategic and user-focused approach from the outset, aligning use cases, integrating data from other departments and streamlining processes for providing the data as free-text insights. This would have promoted broader application and adoption of the tool, maximizing its overall value. On the other hand, we have also experienced multiple examples of companies where similar use cases are being addressed in silos across different departments. This not only results in waste of resources, drives cost, and creates internal political “battles”, but also hinders adoption by creating an unnecessarily complex AI tool landscape that confuses end-users.