The fast-paced world of Fast Moving Consumer Goods (FMCG) relies heavily on innovation to stay competitive. From household essentials to personal care favorites, the FMCG industry is in a perpetual state of evolution, continuously adapting to meet the ever-changing demands of global consumers. Central to this evolution is the ongoing digital transformation of Research and Development (R&D) laboratories, Quality Assurance (QA) laboratories and Pilot plants where technology is revolutionizing the product development process from conception to market launch.

This blog is based on a live roundtable discussion among Consumer Goods industry experts which shed light on the pivotal role of digital transformation in Research and Development (R&D) labs (watch the recording here).

In contrast to the heavily regulated pharmaceutical sector, the FMCG industry operates with fewer regulatory constraints. However, it still faces a unique set of challenges. Regulatory requirements fluctuate across countries, and unlike pharmaceutical trials, there are no standardized phases. Thus, the FMCG industry maneuvers through a regulatory landscape tailored to each product variant.
Nonetheless, it often draws inspiration from the innovative solutions pioneered in the pharmaceutical sector to address its digitalization needs.

Digital Imperative

Like many other industries, the FMCG industry aims to shorten the innovation cycle by improving efficiency and enabling in silico design to reduce the number of designed trials. This relies on capturing a maximum amount of data per experiment in a structured way and making it accessible to inform the next experiment, reducing the need for repeating experiments.

“The goal is to ensure that the data is valuable beyond the lifetime of a project and becomes a long-term resource,” explains Bernhard Sonderegger, Group Leader Scientific Applications and Data management at Nestle. 

On an other side, efforts are being made to automate labs and connect instruments for a seamless transfer of data. This would greatly accelerate repetitive process and eliminate potential human errors, both in process and data collection.

“Whatever can be converted to high throughput should be converted to high throughput,” states Amit Chandra, Distinguished Scientist, Driving Botanical Innovation and Strategy at Amway R&D.

“Integration is very important to free us from human errors,” adds Maneesh Sharma, Global R&D Innovation leader at The Clorox Company. 

Simplifying data capture

In the Consumer Goods industry, data capture remains hybrid, a combination of manual and digital capture depending on the situation and scientist. However, the global goal is to go paperless as it will significantly improves traceability, which is crucial for audits.

To facilitate this transition, traditional lab informatics tools such as the Electronic Lab Notebook (ELN) for early development and sample preparation, alongside the Laboratory Information Management System (LIMS) for routine analysis and QC testing, are being widely adopted. Yet, there's unanimous agreement that manual data entry into these systems remains a time-consuming endeavor, hampering productivity (see White Paper Why ELN needs a Helping Hand…or Voice).

“At Clorox, we put the people at the center of what we do,” emphasizes Maneesh Sharma, Global R&D Innovation leader at The Clorox Company. “When we put a data capture process in place, we make sure sure that it is the most convenient and widely accepted process. This is why we are partnering with LabTwin to enable hands-free data entry into our ELN.”

Erdem Akman, Ice Cream R&D Digital Transformation Leader at Unilever, understands the relevance of such a solution as, in the laboratories developing ice creams, time is of the essence due to the rapid melting of ice cream samples and scientists need to hurry to collect data, hindered by their hands often covered with sticky sugary milk.

Such challenges underscore the need for tailored solutions within the Consumer Goods industry, given the diverse and intricate nature of its products, necessitating adaptable approaches to support scientists in their research endeavors.

Improving data accessibility

Collecting data in a digital form is essential to streamline analysis and speed up collaboration. But data is only useful as long as it is available to both train models and inform everyone who needs to make descisions based on it.

“Before, the data was managed during the project, exchanged in emails and PowerPoint presentations, and ended up in a archive,” recalls Bernhard Sonderegger, Group Leader Scientific Applications and Data management at Nestle. “Now we want to build FAIR data assets which can be leveraged when needed”

“The purpose of the data is to make decision, so we want to be able to rely on the data that we capture, to find it and use it,” added Maneesh Sharma, Global R&D Innovation leader at The Clorox Company. “And for this, the system needs to be user-friendly to be used by people from different backgrounds.”

Establishing standardization across sites and departments

Achieving standardization across sites and departments is essential for centralizing and cross-analyzing data to inform decisions effectively. This process begins with ensuring that data is standardized in format and originates from comparable procedures, irrespective of the site or department where it was captured.

Standardization of procedures:

Many Consumer Goods organizations have labs in multiple sites and countries, facing the challenge of site-specific testing methods, often executed using different instruments and procedures, resulting in local variabilities. This diversity poses obstacles to data pooling and reproducibility, prompting concerted efforts to unify procedures globally.

“We need some level of harmonization for consistency,” highlights Amit Chandra, Distinguished Scientist, Driving Botanical Innovation and Strategy at Amway R&D. “Right now, we have harmonized our hardware and we have adopted a cloud-based system to share data. This enhances reproducibility. Now, we are working on automatically loading the instrumentation parameters from the primary to the transfer lab through our connected system. This makes running of the transfer method plug-and-play and prevent errors.”

In addition to standardizing procedures, another significant challenge arises: standardizing scoring methods for sensory assessments using uniform scales. When evaluating product parameters, it is crucial to rate them on a quantitative scale rather than relying solely on qualitative assessments depending on the tester. This quantitative approach ensures consistency and reproducibility across different labs, enhancing the reliability of sensory evaluations.

“We are trying to standardizing the methodology for sensory assessments by developing standard scales and matching consumer feedback with these scales. For example, how much sweet is a very sweet ice cream?” elaborates Erdem Akman, Ice Cream R&D Digital Transformation Leader at Unilever.

Standardization of data

The challenge is even bigger when it encompasses different departments with different requirements, whether from a  regulatory point of view or an operational point of view, resulting in a large variation in the way data is captured and in which format.

“We work in Agricultural Sciences, Food Safety, Health Science, Clinical Trials, Pet Care. But Lab data, for example HPLC data, should be in a standard format, whether it is collected at the Pilot Plant or in the Food Safety department which have different data management systems,” shares Bernhard Sonderegger, Group Leader Scientific Applications and Data management at Nestle. “And sample IDs and terminologies should be unified through an ontology system. In the proteomics lab, they are using UniProt as reference for the name of a protein where in Clinical Trials, they are using diagnostic markers in medical language to identify the same proteins. We cannot expect our specialists in all those fields to agree on one terminology, so we need to have an ontology system embedded in our systems because at the end of the day, it is the people in the labs that own their own data format.”

While promising technologies like data mesh and knowledge graphs offer potential solutions, their full utilization beyond academic research remains limited. Currently, many organizations attempt to enforce standard templates through shared lab informatics platforms to ensure data capture in a consistent format. However, customization is often necessary to accommodate specific requirements.

“In R&D the primary goal is to design something new and different, but you need standard templates, so that’s the common friction point. It is hard to fit both FMCG category Ice Cream and Shampoo in a same ELN template therefore, we always customize some parts,” explains Erdem Akman, Ice Cream R&D Digital Transformation Leader at Unilever.

Addressing the role of AI in this context, Erdem raises important questions about the level of standardization necessary for scientists in the light of the recent progress made in the field. He ponders, "What does AI require in this space? Do we need extensive standardization, or can AI extract insights from unstructured data in the ELN and provide results?". These questions highlight the evolving landscape of data standardization and the potential increasing role of AI in navigating this complexity.

This blog is based a live roundtable discussion among Consumer Goods industry experts which shed light on the pivotal role of digital transformation in Research and Development (R&D) labs (watch the recording here).

If you are interested in discussing your specific challenges and the value of LabTwin digital lab assistant, book a call with one of our experts.