Your daily activities, from the food you eat to the purchases you make, have carbon emissions embedded in them. At Tweak, we track your transactions and categorize them by type, including Household, Food, Goods, and Transport. While we calculate the carbon footprint of food-related services, we exclude services that are out of the users control such as education or healthcare (click here for more information about this). We then use these categorisations to calculate your carbon footprint score, which reflects your impact on the environment.
For each transaction, we multiply the amount you spent in that category by its respective carbon emissions multiplier either following a top down approach (spend-based methodology) or a bottom up approach (product-level methodology) depending on the level of information available.
Our top-down approach relies on the Exiobase (1) dataset, which is a widely used spend-based emission database in the research and policy-making communities. Spend-based assessments estimate emissions by modeling the total money spent in a particular sector of the economy and the impact of trade, providing an average level of emissions per unit of currency. We use this approach to give us an emission by which to multiply your rand amount purchases. This is how we determine the carbon footprint of purchases such as clothing and household items.
For example, if you spend R500 at Mr Price, we calculate your carbon footprint for that purchase by multiplying your transaction amount by the carbon emissions multiplier of the clothing industry in South Africa based on the Exiobase database calculations. However, this only provides an average level of emissions and may not accurately reflect the emissions of individual products or services.
Our bottom-up approach uses product footprint databases such as Our World In Data (2) to calculate emissions based on the Life Cycle Analysis (LCA) of a specific product. This approach enables much more precise emissions estimates, as it considers the specific quantity of a product that was purchased. We use this approach to calculate the carbon footprint of fuel purchases and other products for which product-level data is available.
For example, if you spend R500 at an Engen petrol station, we can calculate your exact carbon footprint by dividing your transaction amount by the rand per liter of petrol amount and multiplying it by the CO2e/Liter of petrol. This approach is more accurate and reliable because it uses the exact amount of product purchased versus an industry average.
In addition to these approaches, we also use your quiz information to add qualitative information to improve the accuracy of the footprint calculation. For example, we give different weighting to your grocery purchases depending on your diet. We also use this information to divide your transaction by the number of people that live in your household or that you buy food for. If we cannot find any electricity transactions, we use the electricity footprint calculated in the quiz. By using both quantitative and qualitative data, we aim to provide a comprehensive and accurate picture of your carbon footprint.
We also assign a confidence score to each transaction category, reflecting the level of accuracy of the estimate. While spend-based data provides valuable insights into an individual's carbon footprint, it has its limitations and challenges, including the fact that it only provides an average level of emissions. Therefore, we may assign a lower confidence score to transaction categories that rely on spend-based data, such as clothing and food items. Conversely, transaction categories that rely on product-level data, such as fuel purchases, will have a higher confidence score.
In summary, our carbon accounting software calculates your carbon footprint score by combining top-down and bottom-up approaches with qualitative information from the quiz. We continually strive to improve the accuracy of our estimates by using the most up-to-date data available and refining our methods.