How Chuw works

How Chuw explains your food

The Chuw Score rates food on a 0–100 scale from its declared nutrition, its ingredient list and how processed it is, and shows the reasons behind every score. Here is how a camera scan becomes a score you can trace.

Step 01

Scan your food — barcode or not

Chuw starts from what is actually on the pack or the plate. Scan a barcode, photograph a label, or point your camera at a meal:

Barcodes & labels

Look a product up by barcode, or photograph its nutrition and ingredient labels. Chuw reads the declared values per 100 g or 100 ml.

Plated meals & recipes

Recognizes home-cooked meals, restaurant dishes and written recipes. Where nutrition has to be estimated, Chuw says so.

Fresh produce

Recognizes fresh, whole foods without labels or packaging.

How to read a food label, with one real pack annotated
Step 02

One score, worked out the same way every time

The Chuw Score compares foods on a common basis: declared nutrition per 100 g or 100 ml, the ingredient list, and how processed the product is. It starts at 100. Fixed rules take points off for what works against a food and add points for what works for it. The same product gets the same score for everyone — your goals never change it.

What lowers the score

  • Ultra-processing (NOVA 4)
  • Sugars, including disguised ones like syrups and juice concentrates
  • Saturated fat and industrial trans fat
  • High in salt
  • Little fibre for the amount of carbohydrate
  • Non-sugar sweeteners, nitrites in cured meat, alcohol and calorie-rich drinks

What raises the score

  • Fibre, at the EU “source of” and “high in” levels
  • A variety of whole plant ingredients
  • Protein for the calories it comes with
  • Recognized omega-3 foods and prebiotic fibre
  • Live cultures and polyphenol-rich ingredients

Built-in limits

  • For processed (NOVA 3) and ultra-processed (NOVA 4) foods, plus points count for half. Extra points can't hide the processing.
  • Salt sets a ceiling. It starts to tighten above 400 mg of sodium per 100 g and reaches 54 at 800 mg, however good the rest looks.
  • A food can earn more than 100 before display. That shows as a gold ring; averages count each item up to 100.
What ultra-processed means, and how to spot it on a labelSee the score on a real recipe: nasi goreng with fried egg

Five bands, shown by shape as well as color

The number maps to a band. Every band has its own shape, so color is never the only signal.

  • PerfectAbove 100 before display
  • Excellent85–100
  • Good70–84
  • In moderation55–69
  • Limit0–54
Step 03

The score is an entry point, not a verdict

A single number helps you compare options, but the reasons matter more. Every result puts them within reach:

  1. 1The score, band and shape at the top.
  2. 2A good point and a watch-out, in plain language.
  3. 3Your goals: for each goal you chose, whether this food supports it.
  4. 4Breakdown tab: every factor behind the score, with a sign for how much it moved the result.
  5. 5Guidance and Science tabs: practical next steps, and the sources behind the rules.

Crunchy honey and oat granola

High in fibreNo artificial colorsSource of protein
63
In moderation

Fibre contribution

Wholegrain oats bring useful fibre.

+

Plant variety

Oats, almonds and sunflower seeds count as three whole plant ingredients.

+

Ultra-processed

Made with industrial ingredients such as additives.

−−

Free-sugar source

Glucose syrup, honey and sugar add free sugars: about 22 g per 100 g.

−−

Fibre vs. carbs

Little fibre for the amount of carbohydrate.

−

Plus points count half

This product is processed, so its plus points (such as fibre or protein) count for half. The whole score is not halved.

−

Nutrient combination

A combination of fat, sugars or carbohydrate with salt adds a small deduction.

−
Final score63

The example is a real breakdown from Chuw's scoring rules. Chuw shows both sides in context, so you can see how the result was reached, then compare, save, or adjust — the choice stays yours.

When Chuw isn't sure, it says so

When the label leaves gaps, Chuw still gives a score. It never fills a gap in the product's favor:

  • If a value is estimated rather than read from a label, Chuw marks the score as estimated and says what would make it more exact, such as a photo of the label. An estimate can lower a score but never earns plus points.
  • Carbohydrates are compared on one basis. When a US-style label counts fibre inside “total carbohydrate”, Chuw subtracts it before judging fibre per carbohydrate.
  • Score limits are named. When salt, saturated fat or sugar keeps a product from the top scores, the breakdown names that factor.
Step 04

Add your focus — the score stays the same

Nutrition is personal, so Chuw lets you choose up to three goals or trackers in Settings. Leave it blank for General Health. Your focus doesn't change the score. It adds a second lens: for each scan, Chuw checks the evidence that relates to your goal.

Goals

  • Gut Health & Digestion

    Plant variety, fibre, prebiotic fibre and live cultures count in its favor. Ultra-processing, sweeteners, maltodextrin and alcohol are flagged.

  • Weight & Energy

    Fibre, plant variety and protein count in its favor. Added sugar, sugary drinks and ultra-processing are flagged. Chuw doesn't predict your blood-sugar response.

  • Heart & Longevity

    Omega-3, fibre and mostly unsaturated fat count in its favor. Salt, saturated fat, red meat, processed meat and cured meat are flagged.

  • Brain health

    Recognized omega-3 foods such as oily fish, walnuts, flax and chia count in its favor. Ultra-processing, high-mercury fish, salt, trans fat and alcohol are flagged.

  • Fitness & Muscle

    Foods with plenty of protein for their calories count in its favor. Added sugar, saturated fat and ultra-processing are flagged. Chuw doesn't infer complete amino-acid coverage or recovery effects.

Tracker

  • Hormones & SkinTracker

    Not a goal: there is no grounded evidence yet for a product-level target. Chuw records food facts that may be relevant and doesn't judge them.

On each result a goal shows one of five statuses: Supports, Concerns, Mixed, No clear match, or Not assessed. Trackers only show context.

Follow your goals in a food diary, without counting calories

You stay in control

  • No must-eat or must-avoid rules
  • Relevant evidence, not prescriptions
  • The reasoning behind every score

Built on evidence, not hype

Chuw is made for everyday questions about food. It is clear about what it knows, honest about uncertainty, respectful of your choices, and careful with your data.

  • No fear or drama: we don't use labels like "toxic," "clean," or "bad." Food is confusing enough without scare tactics.
  • Honest about limitations: if a label is incomplete or a scan is uncertain, Chuw says so clearly.
  • Wellness, not prescriptions: Chuw helps with everyday food choices and leaves the decision with you.

Methodology and limitations

The Chuw Score comes from a versioned scoring model. AI helps read labels, recognize ingredients and explain the evidence; fixed, deterministic rules calculate the score from declared nutrition values per 100 g or 100 ml, ingredients and processing signals. Estimated values are marked as estimated and can lower a score but never raise it. A missing value is never counted as zero. The displayed score uses a 0–100 scale, with a gold ring when the underlying result earns more than 100. The score describes the product evidence Chuw can assess. It is a wellness heuristic, not a clinically validated measurement, a portion recommendation or a prediction of your personal health.

Questions about the score

How is the Chuw score calculated?
Fixed rules start at 100 and take points off or add them, based on the declared nutrition per 100 g or 100 ml, the ingredient list and how processed the food is (NOVA). AI helps read the label and recognize ingredients; it doesn't set the score.
Do my health goals change the score?
No. The same product gets the same score for everyone. Up to three goals add a second lens: for each scan, Chuw checks the evidence that relates to your goal.
What happens when a label is incomplete?
Chuw still gives a score and marks it as estimated. An estimate can lower a score but never earns plus points, and a missing value is never counted as zero.
What do the score bands mean?
85–100 is Excellent, 70–84 Good, 55–69 In moderation and 0–54 Limit. A food that earns more than 100 before display shows a gold ring.
Where do the rules come from?
From public sources: the NOVA classification (Monteiro et al., 2019) for processing, and EU Regulation (EC) No 1924/2006 for nutrition claim levels such as “high in fibre”. The Science tab on each result shows the sources behind the rules.
What to compare when you choose a food scanner app

Ready to understand your food?

No calorie counting. No forbidden foods. No hype.

Free to use. Try it without an account.

Nutrition claim definitions follow EU Regulation (EC) No 1924/2006. Assessments use the label's declared values per 100 g or 100 ml.