Accelerate CMC Development With Team Topologies
By Irwin Hirsh, Q-Specialists AB
In their book Team Topologies, Matthew Skelton and Manuel Pais apply the educational concepts of cognitive load proposed by Sweller1,2 and generate a methodology for the creation of balanced and productive teams of software developers. Perhaps not surprisingly, these approaches to development teams map well to knowledge workers such as CMC development scientists.
Cognitive Load
Three types of “load” are in focus:
- Intrinsic load: the inherent difficulty of the tasks the team needs to perform
- Extraneous load: the way information or tasks are presented to the team, which can either help or hinder their understanding and efficiency
- Germane load: the mental effort required to create new knowledge and skills or to solve problems effectively
Broadly speaking, Skelton and Pais propose that by organizing teams to manage cognitive load across development projects — that is, divide up the type of “standard” work and problem-solving for which each team is held accountable — companies improve efficiency, innovation, and overall performance. I define standard work in a previous article.
Deliberate Design
The authors also provide evidence and arguments affirming Conway’s Law3 that organizational design and how knowledge sharing flows between the teams are highly impactful on the outputs created. Therefore, consciously designed team structures, what they called team topologies, are a critical factor in business success.
For example, a structure enabling specialists to maximize their time focused on creatively solving problems (germane load) creates the most value to a development organization. This can be achieved if the intrinsic load is minimized with automation. A skilled supporting team structure that is intimate with the needs of the specialists can minimize the extraneous load on such specialists by creating a standardized way of working or even a platform that is easy to use and well communicated.
In this article, I present my thoughts on how to intentionally structure CMC development teams into topologies analogous to those suggested by Skelton and Pais. Guidance is given on how one can create more efficient, engaging, and balanced workloads that allow each team to focus on their germane loads.
Next week, I’ll follow up with a discussion about strategy for structuring teams when development activity is also outsourced. The following discussion applies to internally focused development.
Return On Investments (ROI)
When the development teams have increased clarity on their roles and responsibilities and the work standards (SOPs) provide governance on who does what and when, productivity is ready to go parabolic. Investing in clarity is never a waste and, in addition to productivity, employee engagement will be enhanced. Clarity for the worker enables the sense of mastery and autonomy that almost all knowledge workers crave. For an in-depth understanding about the business benefits of clarity and the fallacies of VUCA thinking see Karin Martin’s book Clarity First.4
Time To Competency
Sadly, it is often challenging to find experienced people capable of navigating the intrinsic load, and there is a scarcity of people capable of tackling the germane load with deep experience and well-practiced methodologies. Therefore, time to competency must be minimized through a well-designed training program that leverages independent learning delivered in a timely manner that is near to its use. To achieve this, I recommend adopting concepts as discussed by Robert Brinkerhoff and Anne Apking in their 2001 book High Impact Learning.
Without this type of program, the most experienced workers are disproportionally distracted from where they can add the most value because of training responsibilities (both formal and ad hoc) and a demand at the intrinsic load level, which would otherwise be covered by the neophytes as part of their learn by doing training.
By consciously structuring the CMC development design teams with an efficient and balanced flow of information — as well as a common focus on solving problems to meet the customer needs — we are well along the path to improving efficiency and effectiveness.
Teams, if they are long-lived, will mature and require adjustments. However, a larger benefit can be found by adjusting what are often domain or organogram-focused topologies, with metrics and rewards based upon their silo. Such default structures can be destructive if not reconsidered thoughtfully.
Default structures often result in suboptimization of the final output, in that they meet the needs of the silo but not the needs of the customer (end-user or internal). Metrics that reward the silo, such as compliance with methodologies and achieving inward-facing metrics, are usually the root cause. Think of a call center more concerned about the number of calls completed per FTE per day than the actual value added to the business they support.
Cross-functional teams can be difficult to create because of “political” or geographic constraints. It may be impossible to create a formal team, but it may be very possible to move them physically closer and/or increase communications and knowledge sharing that supports the desired topology.
Take whatever steps necessary to have collaborative time focused on problem-solving and decision-making. Knowledge sharing is essential; people new to projects must come to understand the details and challenges, but such knowledge should by and large be available through digital means that help rapidly on-board team members. Face time is best spent putting knowledge into action through problem-solving (this includes coaching).
CMC Development Teams
Mitigating the Risks of Churn
In an ideal world, teams are stable, long-lived, and have the experience and knowledge to support the work from end to end (at least from early development to product launch). However, there are some very disturbing trends that upset this need for long-lived teams. According to a measure by the U.S. Bureau of Labor Statistics, the trend from 2012 to 2022 across all occupations is toward shorter tenure with any given employer. The bulk of the specialists I meet at customer sites with more than 15 years of experience are usually baby boomers very close to retirement, and my younger LinkedIn connections seem to be continuing the U.S. trend in tenure referenced above.
Having stable development teams that share knowledge well and are highly functional is a competitive advantage. Once teams are built, there are some common and well-tested tools available to encourage longevity in roles and knowledge retention and sharing within the teams. As mentioned above, time-to-competency is paramount for new employees. In some cases, this may simply be understanding the current state of the projects in focus rather than gaps in technical skills. In other situations, it can be both. Collecting this knowledge, digitizing it, and sharing it is discussed in previous articles.
In general (for all the teams), CMC development staff are highly educated, self-motivated, and earning well above a living wage. Therefore, their motivation can best be understood as the need to feel good about the work they are doing and the achievements that they are making in that work. This attitude is best described by Daniel Pink in his book Drive, wherein he discusses the need for three things: mastery, autonomy, and purpose. When mentors and line managers make great efforts to enable technical mastery and working autonomously and the site senior leadership routinely communicates the Simon Sinek “why” of the business, team members are likely to extend their stay. If they don’t, at least there’s a better chance they will maximize their productivity and engagement.
Team Topologies CMC
1. Stream-Aligned Teams
These teams are aligned to a specific flow of work from a customer or business perspective. In the CMC context, these teams would be responsible for end-to-end processes within their domain.
Examples:
- Formulation development team: focuses on developing the drug formulation, ensuring stability, efficacy, and safety
- Analytical development team: develops and validates analytical methods to ensure the quality of the drug substance and drug product
- Process development team: works on scaling up the manufacturing process from lab scale or medicinal chemistry route to pilot and finally the commercial scale
2. Enabling Teams
These teams help stream-aligned teams to overcome obstacles and acquire missing capabilities. They often work temporarily with stream-aligned teams to transfer knowledge and skills.
Examples:
- Regulatory affairs team: provides guidance on regulatory requirements and helps prepare documentation for regulatory submissions. They must be able to guide the development team to deliver the data that supports the authoring of the dossier. This is a life cycle activity that peaks at the submission of the NME to the health authorities.
- Quality assurance team: ensures compliance with good manufacturing practices (GMP) and other quality standards. They must support the development team in understanding and performing quality risk management (QRM), from facilitating risk assessments to determining potential CQAs or the impact of process changes on product safety. When appropriate, they must also take part in mitigation activities, such as, for example, supplier qualification or CAPA assistance.
- Technology transfer team: facilitates the transfer of technology and processes from development to manufacturing sites. Often, the tech transfer responsibility is in the MSAT group. Their support is fundamental to understanding the gaps due to changes in scale, location, and equipment. Their input as SMEs into process failure mode and effects analysis (FMEA) is fundamental to a successful transfer, even internally.
In addition to guidance on working in compliance appropriate for the phase of development, enabling teams must be able to support what is commonly called quality by design. That is, they help the stream-aligned teams keep focus on the capability of the process and routinely prove, with data, the current capability and gaps in capability of the process to deliver a product meeting the currently defined product quality attributes.
The tech transfer team may have to draw on expertise from the complicated-subsystem teams (described below) to understand unit operations on a detailed level that supports the identification of critical control points, such as, for example, using fluid dynamics modeling (CFD) to resolve differences in a stirred tank design at the CMO, which is very different from the pilot-scale in-house design.
3. Complicated-subsystem Teams
These teams are responsible for areas that require deep specialist knowledge. They handle complex subsystems that are critical to the overall system but are not the primary focus of stream-aligned teams.
Examples:
- Bioprocess engineering team: specializes in the development and optimization of biotechnological processes, such as fermentation or cell culture
- Material science team: focuses on the properties and behaviors of materials used in drug formulation and packaging
- Automation and control team: develops and maintains automated systems for manufacturing processes, ensuring precision and efficiency
- Modeling group: This new specialist group creates mathematical, statistical, and machine learning models, supporting decision-making and accelerating time to market. Their models need to be integrated into the decision-making process and shared across teams to optimize performance and outcomes.
4. Platform Teams
These teams provide foundational services and tools that other teams can use to deliver work more efficiently. They create and maintain platforms that reduce cognitive load for stream-aligned teams.
Examples:
- Manufacturing IT systems team: manages the IT infrastructure and software systems used in manufacturing, such as manufacturing execution systems (MES) and laboratory information management systems (LIMS)
- Supply chain engagement team: ensures the availability of raw materials and coordinates logistics to support manufacturing operations
- Facilities and equipment team: maintains the physical infrastructure and equipment used in manufacturing, ensuring they are in good working order and compliant with regulations
- Modeling group: develops and maintains mathematical, statistical, and machine learning models, providing essential insights and predictions that support decision-making and speed up time to market. They ensure that model outputs are actionable and integrated into the broader knowledge management system.
References:
- Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12(2), 257-285.
- Sweller, J. (1994). Cognitive load theory, learning difficulty, and instructional design. Learning and Instruction, 4(4), 295-312.
- Conway, M. E. (1968). How do committees invent? Datamation, 14(4), 28-31
- Martin, K (2018). Clarity First: How Smart Leaders and Organizations Achieve Outstanding Performance, New York: McGraw-Hill Education. ISBN 978-1259837357
About the Author:
Irwin Hirsh has nearly 30 years of pharma experience with a background in CMC encompassing discovery, development, manufacturing, quality systems, QRM, and process validation. In 2008, Irwin joined Novo Nordisk, focusing on quality roles and spearheading initiatives related to QRM and life cycle approaches to validation. Subsequently, he transitioned to the Merck (DE) Healthcare division, where he held director roles within the biosimilars and biopharma business units. In 2018, he became a consultant concentrating on enhancing business efficiency and effectiveness. His primary focus involves building process-oriented systems within CMC and quality departments along with implementing digital tools for knowledge management and sharing.