Modern diets are increasingly rich in processed products. Almost two thirds of the food supply consist of packaged foods but, until now, most national food composition datasets have focused on generic foods that cannot reflect the unique characteristics of branded products. For instance, in the market, the term “Granola” can refer to a wide range of products with vastly different nutritional characteristics, as recipes and ingredients vary significantly across brands. In this context, the need for food composition datasets that can provide detailed information about the current foodscape has driven an international initiative for branded food composition databases (BFCDs).

For the most part, BFCDs are based on food labels generated by manufacturers. In European Union, Regulation (EC) 1169/2011 defines the general principles, requirements, and responsibilities related to information for consumers, or food labelling. Food labels must mention mandatory information, such as ingredients, macronutrients, and allergens but can also provide non-mandatory information such as nutrition and health claims, quality schemes, etc. Thus, food labels can be a valuable source of not only food composition data but also of a wide range of other information.

BFCDs offer comprehensive data for individual products, complementing the information found in national food composition datasets. In the market, this specificity supports consumers in making informed dietary choices and enables health professionals to deliver tailored nutrition advice. Researchers benefit from detailed data on individual food items, facilitating food, nutrition, and health studies. Additionally, food manufacturers can use these datasets for product development and reformulation, as well as for marketing purposes. BFCDs also enable monitoring of changes in the market over time and can be valuable for policymakers. Detailed and reliable data on branded foods can also enhance food safety, helping trace potential allergens and contaminants.

Although BFCDs are increasingly in demand, their development and maintenance present several challenges. Data collection for representative numbers of foods from various manufacturers requires substantial effort and resources, regardless of the methods used (e.g., supermarket data collection, web scraping, crowdsourcing, or obtaining data directly from manufacturers). While collaboration with manufacturers is often the most reliable method, food companies typically hesitate to share their data. Additionally, the need to keep data up to date is continuous, as branded products are subject to rapid changes due to reformulation, launch and withdrawal, or other modifications. Data harmonisation among different brands requires strict standardisation in methodology processes. Also, developing a robust, user-friendly platform that can handle large volumes of data and provide easy access and use for stakeholders require significant technical expertise and ongoing resources.

Food composition datasets are changing to meet new demands and branded food datasets can offer significant benefits for stakeholders, supporting informed dietary choices, personalised nutrition advice, and effective food safety protocols. However, developing and maintaining these databases is challenging due to extensive data collection needs, the need for constant updates, and manufacturers’ reluctance to share information. EuroFIR recognises the need for branded data and is working with Members to enhance understanding and facilitate delivery of these databases through a number of different mechanisms.