eCommerce Widget for Nutrition and Sustainability
Leveraging personal eCommerce widget to support decision making towards healthier and more sustainable food choices.
Diet-related non-communicable diseases have become the leading cause of mortality globally, accounting for more deaths than non-diet-related mortality causes combined. In addition, the production and consumption of food is one of the biggest contributors to greenhouse gas emissions. We believe that digital personal assistants can support users in making health-beneficial and more sustainable decisions in our daily lives. Motivated through the advances of related technology, we contribute to this field by piloting a research study on autonomous widgets, i.e. browser extensions that extend existing eCommerce websites to support consumers in their decision making.
The research project ‘eCommerce Widget for Nutrition and Sustainability’ aims to assess the potential of digital interventions to improve individual food selection in eCommerce environments. In particular, the goal of this research project is to identify the impact of nutrition-oriented (e.g. Nutri-Score) and sustainability-oriented (e.g. Beelong) labels on consumer behavior. Therefore, we will implement a web-based browser extension that seamlessly integrates into existing online grocery shopping websites to visualize the nutritional and/or environmental impact of a product to the consumer. Further, the application is designed to track how consumers engage with the website (e.g. which products are looked at and which products are actually purchased) post-exposure to either nutritional and/or sustainability labels.
This project is based on product data from the eCommerce website as well as data from the AutoID Eatfit database, which includes more than 50k products with data on nutrients and ingredients for food products sold in Switzerland. This database enables us to display relevant information, such as Nutri-Score, as well as information related to sustainability, such as country of origin, ec0-labels, packaging etc.
We plan to recruit between 100 and 1000 volunteers for the eCommerce shopping study within the university laboratories available at ETH, HSG and FAU. After collecting behavior data from the experiments, we will analyse the user behavior as a response to label exposure using descriptive statistics and machine learning to detect if and how such labels can influence customer decision-making during shopping journeys in eCommerce environments.