How we calculate a product score
FoodScan gives packaged foods a score from 0 to 100. The goal is to help you understand a product faster, compare options more easily, and notice ingredients or nutrition signals that are easy to miss on a label.
This is not medical advice. Scores can be inaccurate, incomplete or outdated. Always check the physical package, especially for allergens, medical diets, pregnancy, children, chronic conditions or any health decision.
What the score means
A quick reading of a packaged food, not a final judgment.
The FoodScan score is a simple number between 0 and 100. A higher score usually means the product has a better nutritional profile, fewer concerning ingredients, and a lower level of processing.
The score is meant to be practical. It helps you understand a product in a few seconds while you are shopping. It can also help you compare two similar products, for example two cereals, two yogurts, two sauces or two protein bars.
It should not be treated as a personal recommendation. A product with a high score is not automatically right for you, and a product with a low score is not automatically forbidden. Your allergies, your health history, your portion size, your diet and your personal goals all matter.
Fast signal, deeper details below
The score is only the summary. In the app, we also show the reasons behind it, including sugar, salt, fat, additives, processing level, allergens, labels and missing data warnings when needed.
The 0 to 100 scale
The same color logic is used across the product page.
We use a 0 to 100 scale because it is easy to read quickly. The color is there to help you scan the result faster, but the explanation below the score is more important than the number alone.
A score is not a moral label. It does not mean you should never eat a product. It simply gives a quick view of how the product looks based on the information available to us.
How the score is calculated
We combine nutrition, ingredients, processing and data quality.
The score is built from several signals. Some signals can improve the score, such as fiber or protein. Others can reduce it, such as high sugar, high salt, high saturated fat, certain additives or a high processing level.
We identify the product
When a barcode or product photo is available, we look for the product name, brand, category, serving information, nutrition facts and ingredient list.
We read the nutrition facts
We evaluate calories, sugar, salt or sodium, fat, saturated fat, fiber and protein. When possible, we compare values per 100 g or 100 ml to make products easier to compare.
We analyze ingredients and additives
We look for additives, sweeteners, preservatives, colorants, oils, palm oil, allergens and ingredient signals that may matter for the product explanation.
We estimate processing level
When the ingredient list allows it, we use processing indicators inspired by the NOVA classification to understand whether the product is closer to whole food or ultra-processed food.
We combine the signals
The final score is calculated from the available signals. If important data is missing, the product page should show that the analysis may be less reliable.
Positive signals
Fiber, protein, simple ingredient lists, low sugar, low salt, low saturated fat and low processing can improve the score.
Negative signals
High sugar, high salt, high saturated fat, some additives, ultra-processing and poor data quality can reduce the score.
Nutrition signals
How we read sugar, salt, fat, fiber and protein.
We use nutrition values mainly per 100 g or 100 ml because serving sizes can vary a lot between brands. This makes comparisons more stable. When the app shows a serving view, it should still make clear that the score is mostly based on standardized values.
Fiber and protein are treated differently. In most packaged foods, more fiber is generally a positive signal. Protein can also be positive, but it does not automatically make a product healthy. A high-protein snack can still be high in sugar, saturated fat, sodium or additives.
Processing level
Why ingredients matter beyond the nutrition table.
A nutrition table does not tell the whole story. Two products can have similar calories, sugar or fat, while one is made from simple ingredients and the other is built from many industrial ingredients. This is why we also look at processing.
When a product looks ultra-processed, the score can be reduced. This usually happens when the ingredient list contains multiple markers such as industrial sweeteners, emulsifiers, modified starches, flavoring agents, colorants, texturizers or long ingredient lists that are not usually found in home cooking.
Processing is not always easy to classify perfectly. Some products sit in a grey zone. When the ingredient list is missing, unclear or translated badly, the processing estimate may be wrong.
Additives and ingredient flags
How we explain E-numbers, sweeteners, preservatives and colorants.
FoodScan detects additives listed in the ingredients and groups them into simple labels. These labels are designed to help users understand what they are looking at. They are not a toxicology report and they do not replace official safety authorities.
Ingredients we consider concerning
Examples can include certain colorants, nitrites in processed meats, or additives that are often discussed in public health guidance. These can reduce the score more strongly.
Ingredients to keep moderate
These are not automatically dangerous, but they may be common in heavily processed foods or worth limiting depending on your diet and frequency of consumption.
Ingredients with no major concern at normal food levels
Some additives, such as ascorbic acid or citric acid, are generally considered low concern when used in typical food amounts.
Additive classifications can evolve. When public information changes, or when we improve our own classification, the explanation for a product may change too.
Where the data comes from
The score depends on the quality of the product record.
FoodScan combines several data sources. This helps us show a useful result even when one source is incomplete, but it also means the analysis can contain mistakes.
Public product data
We may use public and collaborative food databases for barcodes, product names, ingredients, nutrition facts, labels and images.
User feedback and corrections
When users report wrong data, missing ingredients or outdated packaging, we can use those corrections to improve future scans.
Our own database and rules
We maintain our own classification for additives, ingredient flags, score logic, label interpretation and product explanations.
Public nutrition guidance
We use public nutrition thresholds and food labeling logic to help classify nutrients as low, moderate or high.
The product packaging remains the most reliable source. If FoodScan and the physical label disagree, use the physical label.
Where the score has limits
The score is useful, but it cannot understand your full context.
A food score can never fully replace your own judgment. It cannot know your medical history, your allergies, your portion size, your goals, your medications, your pregnancy status, your age or what you eat during the rest of the day.
- The score may be wrong if the product data is incomplete, outdated or entered incorrectly.
- The ingredient list may change before the public product record is updated.
- Allergen information can be incomplete, mistranslated or missing.
- Serving sizes can make a product look better or worse than it really is.
- A high score does not mean a product is suitable for every diet.
- A low score does not mean a product is dangerous in every situation.
- The score does not replace advice from a doctor, dietitian, nutritionist or pharmacist.
Medical and legal notice
FoodScan is provided for general information about packaged foods. It is designed to help users read products more easily. It is not designed to diagnose, treat, prevent or cure any disease.
Information only
The product score, ingredient analysis, additive labels and nutrition comments are informational. They are not medical advice, dietary advice or a personalized health recommendation.
Data can be inaccurate
Product data can be incomplete, incorrect or outdated. Ingredients, allergens, nutrition facts, labels and images may not match the current packaging. Always verify the physical product before buying or consuming it.
Do not use for medical decisions
If you have allergies, diabetes, kidney disease, heart disease, an eating disorder, a chronic condition, are pregnant or breastfeeding, are buying food for a child, or follow a medically restricted diet, speak with a qualified professional. Do not start, stop or change a diet based only on FoodScan.
By using FoodScan, you remain responsible for your food choices. To the fullest extent allowed by law, FoodScan and its publisher do not accept liability for decisions made solely from the score or analysis shown in the app.
You can report incorrect product data from inside the app. If you contributed data and want to request access, correction or deletion where applicable, please refer to our Privacy Policy.