Methodology
How seCO2ia calculates your carbon footprint
seCO2ia turns your everyday purchases into a CO2 estimate using recognised public emission databases — chiefly ADEME and Agribalyse. Here's exactly how we calculate it, and how honest we are about its limits.
How does seCO2ia calculate the CO2 of a product?
For each item on a receipt or barcode, our AI identifies the product type, estimates its weight and reads its price. We then apply a transparent formula: CO2 = estimated weight × emission factor for that product type. For categories where value matters more than weight (e.g. fashion), we use an emission factor per euro spent instead.
Where do the carbon factors come from?
From recognised public databases, not guesses: ADEME Base Carbone 2024 and Agribalyse 3.2, the reference databases of France's ecological-transition agency. For specific cases we also use institutional and peer-reviewed sources (MIT, Eurostat, EPA, DEFRA).
Sources: ADEME Base Carbone · Agribalyse
How reliable are the figures?
They are estimates, not certified measurements — and we say so. Two sources of uncertainty: the AI estimates a product's weight without physically weighing it, and each factor is an average for a product type. So treat a scan as a reliable order of magnitude to guide your choices, not a laboratory result. Every result carries a confidence status, and accuracy improves continuously as the data and AI get better.
How does the community make it more accurate?
Anyone can flag a wrong product, correct its category, weight or name, and vote on others' suggestions. After 5 confirming votes, a correction is applied for everyone — so accuracy compounds with the community.
Does the app itself pollute?
About 22 g of CO2e per year for an active user — negligible next to what it helps you avoid. In a single shopping trip, one better choice already dwarfs it: swapping just one beef steak for chicken saves several kilograms of CO2, far more than the app emits in a whole year.