It's all in the details

Sustained currently focuses on the production stage of the food life cycle in assessing the impact of each product by considering the ingredients of each product.

We use data to assess the estimated environmental impact of the production of food products using their ingredients. This is done using a scientific method called Life Cycle Assessment (LCA) and some background data on the ingredients environmental impact to calculate the impact of a product’s production across a number of environmental impact categories.

Several LCA databases (see full list below) are used in conjunction with a global assessment framework to provide a baseline impact assessment. This is then used by the Sustained systems to determine a relative impact rating for each product.

When it comes to framework, we have recently switched from using the Recipe2016 framework to what is soon to become a European standard, the Product Environmental Footprint (PEF) framework. With this method, impacts are estimated for 16 categories, which we have grouped into 6 broader ones to make sure you don’t need a PhD to understand what we’re telling you!

Impact categories currently assessed:

These calculated impacts are the core inputs into our environmental impact rating calculations and algorithms. Following the PEF framework, the next step after calculating the absolute impacts for each of the 16 categories, is to apply normalisation factors. These allow for apple-to-apple comparison (i.e. compare the categories to each other), by reporting impacts as per person equivalent.

Once normalised, the scores are then weighted to give a better representation of the most important vs. the least important impact categories. This is where PEF becomes very interesting. Stemming from the work of many LCA specialists, and conversations with people from all around the world to understand what matters more to them, the suggested weightings reflect the best consensus available today as to what is most or least important when considering our environment.

These weighted scores across the 16 categories are then added together to assign each product a final score which our assessment system then converts into a rating (A to G) where A is great and G is bad. This rating is relative to all the other products in our system and as food production systems improve, newer products will be rated better than the older ones.

Why relative, you ask? Simply put, food production processes, including farming, pesticides, transport, and more, will continue evolving and hopefully becoming better with less environmental impact over time. Thus our assessment of product impact must adjust relative to the state of the art as this happens, rather than being based on an arbitrary state as of today.

Please note that the rating is estimated due to the nature of the LCA data sets in use. We also do not currently facilitate a full PEF study as this would need to be done using only primary data to perform the assessment (which only manufacturers have access to). Sustained aims to add support for verified ratings over time.

Since we are using ingredients to score products, the availability of composition information from brands is crucial to the accuracy of our estimations. This is why we’ve introduced the concept of a data availability indicator, being transparent with you about how much of the composition we know from the brand, and how much we had to estimate using our own models.

While we want you to know that some brands are more transparent than others, we are not asking you to make sense of how that impacts a product’s overall score. It would go against our main goal, to help you make sustainable choices easily. This is why we’ve baked in a penalty in our ratings to account for the added uncertainty about a product.

Sometimes we have additional information about a product to help you make even more informed choices. These include "Low CO2 emissions" (less climate change impact!), "Local product" (made closer to where you live, so less travel!) and “Seasonal food” (In season now, so less storage, travel and pollution!). These show up right next to the product rating as icons.

The Sustained system data sources in use today:
‍
• Ecoinvent 3.8
• Agribalyse 3.01

The LCA framework used to calculate the product ingredient impact baselines:
‍
• Product Environmental Footprint (PEF)

Currently supported UK stores:

• Sainsbury’s (www.sainsburys.co.uk)
• Ocado (www.ocado.com)
• Tesco (www.tesco.com)

Please note that products from Waitrose, ASDA and other stores are also available in our mobile app and that rating coverage will increase over time as Sustained adds additional databases, ingredients and LCA impacts to the system. Also note that we are not affiliated with any of those stores and our ratings are independently calculated to give you an unbiased view of the world.

Known limitations:

• Product ratings are currently normalised based on scoring 1kg of each ingredient, and do not factor in actual weight of the product. For multi-ingredient products, proportions of each ingredient (%) are then used to derive scores.
• Transport and packaging is included in the impact calculations on a per ingredient basis, but not as the final product production itself.
• Country of origin is not yet included in the impact calculations.
• In the absence of primary data (from the brands and retailers) about how a given tomato is actually being grown, default agricultural processes are assumed in calculating impacts. These defaults have been derived from a number of scientific studies of the impact of ingredients on the environment.

If you have any questions on the methodology, approach, data sources, and scoring, please get in touch directly with us on hello@sustained.com.