The Challenges of Scaling Digital Advances in Life Sciences

Scaling innovations in R&D can be considered challenging in two dimensions: from lab to lab; and from the lab to the end customer. A 100x or 1000x advance in R&D efficiency challenges an organization to adapt, from R&D to engineering, manufacturing, and to the marketplace, where there is a significant business opportunity.

 

Incremental Advances Versus Created Possibilities       

Many of the improvements in life sciences R&D labs today come through introducing digital technologies to existing processes, with an evolution that provides measurable, incremental improvements. These are often delivered by small, agile, highly technical teams, with improvements the organization well understands how to scale. 

However, some of these small teams are realizing the possibilities of orders of magnitude improvements in R&D results – what we’ll call ‘applied digital innovation’. Scaling that magic to the wider organization is another challenge altogether. 

Examples of applied digital innovation have the potential to set a company on a journey for drug developments in months (not decades), costs in millions (not billions), vaccines in months (not years). The benefits of realizing these possibilities are significant to both patients (‘customers’) and shareholders, a dream combination for the industry. This value combination makes a compelling case to invest in digital capability and the organizational transformation necessary to scale newly realized possibilities.

The real challenge for management is how to scale orders of magnitude R&D lab advances across the organization.

The Challenge of Scaling the Large Scale 

Scaling significant innovations in R&D lab performance can be considered challenging in two dimensions. The first is from lab to lab, oftentimes in different countries, with different IT infrastructures, organizational cultures and more. The second is from the lab to the end customer. A 100x or 1000x advance in efficiency or quality delivered by the R&D lab challenges the rest of the organization to adapt to the new reality. This chain is from R&D to engineering, manufacturing, and to the marketplace, where there is a significant business opportunity that can include entirely new business models. 

Our experience suggests the issue of scaling digital innovation across organizations is challenging because much of its current application is myopic. Small teams, working on standalone projects, are very capable of accessing and applying the latest advances in digital technologies to realize possibilities. These teams invariably combine deep domain expertise with an equivalent in advanced scientific software and computing, either within the team or, more often, using external resources.

While the processes and workflows of these teams are revolutionized, adapting the interdependent workflows of the surrounding organization to exploit the advance is a very different challenge. The result of this is often frustration felt by internal teams. The reality has been that organizations always will advance at a slower pace than individual teams. However, the possibilities R&D lab teams can create from digital capabilities demand that accepted reality be challenged. 

One of the best examples is when R&D labs utilize new, automated measurements and data ingestion, then apply AI/machine learning techniques, which deliver final or near final results. Scientists who have accelerated lab delivery work to help the organization adapt and realize its full business potential for customers. There are numerous examples of such successes, from MIT using AI to discover a new antibiotic, to Boston General using image analysis to detect ovulation pattern changes

A significant challenge in scaling digital capabilities across the larger organization is where management must focus. McKinsey takes a broad organizational view of the potential from digital technologies and skills, outlining 10 key ‘battlegrounds’ that represent significant opportunities for value creation within life sciences. McKinsey argues that digital success comes from picking your battles and focusing on “entire parts of the business system, rather than use-by-use cases.” 

The 10 life sciences “battlegrounds” defined by McKinsey include: 

  • Disease understanding
  • Disruptive product design
  • Integrated evidence generation
  • Operational excellence in development
  • Medical impact
  • Precise, real-time customer and patient engagement
  • Industry 4.0
  • New digital and data-drive businesses and business models 
  • Digital organization
  • Technology modernization 

Each battleground is “… an area of the business system where it is possible to deliver value at scale through a “platform solution,” whereby specific data sets, data and analytics platforms, analytical models, and digital experiences for customers and end-users are brought to bear upon a cluster of closely related digital and analytics use cases.” 

McKinsey continues, “The companies that get ahead will therefore be those that seek opportunities to advance their platform plays continuously and find the next generation of solutions before others do.” 

Employing the right “platform solution” to which McKinsey refers is a major challenge. Addressing this challenge is helped greatly by breaking down what is required into three main building blocks. This allows clearer framing for what the solution looks like:

  1. Corporate infrastructure, in particular cloud or on-premises, and associated data management.
  2. Generic/open-source software technology that enables advanced computing (dashboards and analysis, among others).
  3. Customized infrastructure to enable high-value software solutions used by experts to function interdependently with the corporate infrastructure.

Built on top of this ‘platform’ will be specific high-value scientific software solutions, including AI/machine learning tools. This is where applied digital innovation will take place and where possibilities are realized. A key feature of these advanced software solutions will be to make AI/machine learning capabilities readily accessible to scientists. 

Scaling and Senior Leadership

Scaling digital innovation across the organization requires taking the long view, with leadership that is committed to providing governance and sponsoring digital innovation initiatives. Chris Llewelln, Senior Partner at McKinsey in London, states, “Without such sponsorship and a narrative that makes clear the need for change, most business leaders will stay focused on short-term profit and loss targets, lacking any incentive to do things differently.” 

These are the incremental advances from digital technologies identified earlier that the organization knows how to scale. 

Digital strategy should include objectives that encourage the R&D lab to pursue possibilities, not just incremental advances. Llewelln continues, “It cannot be assumed that if investments in people, data, and technology are made, value will follow. A business case must be made at the outset, then reviewed at least quarterly to ensure the link to value holds firm.”

Making a convincing business case for possibilities through lab R&D is much more challenging than one for incremental advances, which typically offer line of sight to value in the next two quarters. 

Earlier work on a practical framework for evaluating an organization’s ability to scale, both for incremental advances and possibilities, can be found in the Enthought five-element model for Digital Maturity: Digital Strategy, Digital Skills, Digital Tools, Data & Data Flow, and Data Infrastructure.

Scaling Takes Time   

Digital transformation is a decade long journey. Scaling digital innovation across an organization is also multi-year, requiring dedication, resources, and focus. Llwellyn writes, “… companies should ask in which battlegrounds they are uniquely positioned to place big bets and which offer the best value for their portfolios. It is unlikely they would be able to transform the business in more than a few battlegrounds within a two-year period, as they would lack the necessary leadership capacity, change readiness, assets, and capabilities.”

And the bar is always moving. Digital technology arguably advances in line with Moore’s Law (the number of transistors on a chip), doubling capabilities roughly every two years. It is critical that science-driven organizations be able to exploit these advances as they become available, whether in their enabling ‘platform’ or the advanced software solutions that will run on it. 

Llewellyn continues, “Cutting-edge analytics techniques such as deep learning, transfer learning, and reinforcement learning will disrupt areas that have already been transformed by advanced analytics…And cognitive computing and natural language generation can drive further productivity advances in business processes that have been automated with robotic process automation.”

Momentum Matters

Delivering early business value through initiatives targeting digital innovation is critical. Without an early win, teams and organizations can lose interest, and leadership moves on to focus on other projects, tools, and initiatives that provide clear value. These projects, tools and initiatives often provide only incremental improvements, measured quarterly, whilst the possibilities enabled by advanced digital capabilities are missed. 

When developing plans to scale applied digital innovation in an R&D lab, think early value – delivering critical wins quickly – with ongoing successes throughout the main project. However, within that path, make structured efforts to discover possibilities, ones that progress in parallel with advancing digital capabilities. The early delivery of value is necessary to secure commensurate investment and commitment from management.

One such example of an early win in materials science, where active learning improved scale-up efficiency in a project in which an R&D lab cut costs by $90,000 per formulation and increased time-to-market 3x. This is a good example of early business value, beyond the incremental, providing momentum and motivation in a multi-year digital transformation initiative.  

In Summary 

Llwellyn sums it up well. “Life-sciences companies are on the cusp of a digital revolution. There are wondrous opportunities to leverage digital and analytics to address significant unmet patient need if these technologies and capabilities can be deployed at scale.”

The life sciences digital revolution Llwellyn refers to, and that must be scaled if it is to succeed, ultimately starts in the R&D lab, with scientists. And those scientists must be equipped with the power of the rapidly advancing scientific software tools and techniques.

Share this article:

Related Content

R&D イノベーションサミット2024「研究開発におけるAIの大規模活用に向けて – デジタル環境で勝ち残る研究開発組織への変革」開催レポート

去る2024年5月30日に、近年注目のAIの大規模活用をテーマに、エンソート主催のプライベートイベントがミッドタウン日比谷6FのBASE Qで開催されました。

Read More

科学研究開発における小規模データの最大活用

多くの伝統的なイノベーション主導の組織では、科学データは特定の短期的な研究質問に答えるために生成され、その後は知的財産を保護するためにアーカイブされます。しかし、将来的にデータを再利用して他の関連する質問に活用することにはあまり注意が払われません。

Read More

科学研究開発リーダーが知っておくべき AI 概念トップ 10

近年のAIのダイナミックな環境で、R&Dリーダーや科学者が、企業の将来を見据えたデータ戦略をより効果的に開発し、画期的な発見に向けて先導していくためには、重要なAIの概念を理解することが不可欠です。

Read More

科学における大規模言語モデルの重要性

OpenAIのChatGPTやGoogleのBardなど、大規模言語モデル(LLM)は自然言語で人と対話する能力において著しい進歩を遂げました。 ユーザーが言葉で要望を入力すれば、LLMは「理解」し、適切な回答を返してくれます。

Read More

ライフサイエンス分野におけるデジタル化拡大の課題

研究開発におけるイノベーションの規模拡大は、ラボか…

Read More

ITは科学の成功にいかに寄与するか

科学と工学の分野においてAIと機械学習の重要性が高まるなか、企業が革新的であるためには、研究開発部門とIT部門のリーダーシップが上手く連携を取ることが重要になっています。予算やポリシー、ベンダー選択が不適切だと、重要な研究プログラムが不必要に阻害されることがあります。また反対に、「なんでもあり」という姿勢が貴重なリソースを浪費したり、組織を新たなセキュリティ上の脅威にさらしたりすることもあります。

Read More

Top 5 Takeaways from the American Chemical Society (ACS) 2023 Fall Meeting: R&D Data, Generative AI and More

By Mike Heiber, Ph.D., Di…

Read More

Life Sciences Labs Optimize with New Digital Technologies and Upskilling

Labs are resetting the tr…

Read More

From Data to Discovery: Exploring the Potential of Generative Models in Materials Informatics Solutions

Generative models can be used in many more areas than just language generation, with one particularly promising area: molecule generation for chemical product development.

Read More

The Importance of Large Language Models in Science Even If You Don’t Work With Language

OpenAI's ChatGPT, Google's Bard, and other similar Large Language Models (LLMs) have made dramatic strides in their ability to interact with people using natural language....

Read More