How OK Calculates Health Scores

OK is a grocery and wellness shopping app for Indian consumers. Every product in OK shows an overall Health Score (0–100) that summarises how nutritionally appropriate the product is for an average adult, computed from information printed on the product's own packaging. This page explains, in detail, how that number is produced, what sources we rely on, and the limits of what it can tell you.

OK does not provide medical advice, diagnosis, or treatment. The score described below is an educational signal to help users compare packaged foods. For any clinical decision — including dietary changes related to a medical condition, pregnancy, allergies, medication interactions, or paediatric nutrition — please consult a qualified healthcare professional. See the Disclaimers section at the bottom of this page.

1. What data we use

Every score on OK is computed from information printed on the product's own packaging. Specifically:

1. Nutrition Information panel — the per-serving and/or per-100 g (or per-100 ml) values for energy, protein, carbohydrates, total sugars, added sugar, dietary fibre, total fat, saturated fat, trans fat, cholesterol, sodium, and any micronutrients the brand has declared.
2. Ingredients list — the full ingredient declaration in the order printed by the manufacturer, including sub-ingredients shown in brackets, and the percentage composition where the brand has declared it.
3. Front-of-pack claims and regulatory marks — brand, product name, FSSAI licence number, manufacturer/marketer information, vegetarian / non-vegetarian mark, allergen declarations, and any nutrition or health claims printed by the manufacturer (e.g. "high in fibre", "no added sugar", "source of protein").
4. Storage, usage and shelf-life information as printed.

We do not use lab tests, third-party nutrition reformulations, or our own opinion of a product. If a value is not printed on the pack, it is not used in the score.

The extraction step is performed by a vision-language AI model on photos of the product label. The structured output is then stored, reviewed, and used as the input to the scoring formula described below.

2. The Health Score (0–100)

The Health Score is a single number meant to represent how nutritionally appropriate a product is for an average adult, judged purely from the information on its own packaging. The same product shows the same score to every user.

The score weighs three components:

2.1 Nutrition profile

Calculated from the per-100 g (or per-100 ml) panel where available, falling back to per-serving values. We consider:

• Energy density (kcal per 100 g)
• Added sugar
• Sodium
• Saturated and trans fat
• Dietary fibre
• Protein density

We use the following reference thresholds, which are drawn from WHO and FSSAI guidance for the general adult population:

2.2 Ingredient quality

A short ingredient list dominated by whole foods scores higher than a long list dominated by refined oils, refined flours, hydrogenated fats, added sugars, or non-nutritive sweeteners.

2.3 Processing level

Products positioned higher on the NOVA processing classification (ultra-processed) carry a penalty. NOVA is published by the University of São Paulo and adopted in FAO guidance [7].

The three components are combined into a single 0–100 score and rendered in the app as:

• 80–100 Excellent
• 60–79 Good
• 40–59 Average
• 20–39 Below Average
• 0–19 Poor

Allergens declared by the user during onboarding are matched against the allergen tags extracted from each product's ingredients and allergen declaration. If there is an overlap, the product is shown with an allergen warning before checkout. This is a binary match against the printed label — it is not a clinical allergy assessment.

3. Daily calorie and macronutrient targets

For users who provide age, sex, height, weight, and activity level, OK computes a daily calorie and macronutrient target. This is used only to contextualise how a product fits into a day's intake.

• Basal Metabolic Rate (BMR) is computed using the Mifflin–St Jeor equation [8], which is the equation recommended by the Academy of Nutrition and Dietetics for non-clinical estimation of resting energy expenditure in healthy adults aged 19–78.
• Total Daily Energy Expenditure (TDEE) = BMR × activity multiplier, using the standard multipliers of 1.2 (sedentary), 1.375 (lightly active), 1.55 (moderately active), 1.725 (very active), and 1.9 (athlete) [9].
• Goal-adjusted calories: muscle gain +300 kcal/day, weight management −250 kcal/day, weight loss −500 kcal/day. These are conservative deltas consistent with general dietetic guidance [10].
• Floors and ceilings: 1200 kcal/day for women and 1500 kcal/day for men are enforced as safety floors; 5000 kcal/day is the ceiling. These thresholds follow NIH/USDA guidance that calorie intake below these floors should only be undertaken under medical supervision.
• Protein target: 1.0 g/kg/day baseline, rising to 1.2–1.8 g/kg/day for active and weight-management goals, in line with the ISSN position stand on protein and exercise [5].
• Other defaults: 25 g/day fibre, ≤ 50 g/day added sugar, ≤ 2300 mg/day sodium, ≤ 20 g/day saturated fat. These align with WHO and FSSAI recommendations [1][2][6].
• When age, height, or weight is missing, we fall back to NFHS-5 (India) population medians and flag the target as an estimate in the UI.

4. Glycemic index and glycemic load

Where a product's pack prints a glycemic index (GI) or glycemic load (GL), we use that value as printed. Otherwise we estimate from the ingredient list and per-serving carbohydrate and fibre values using the standard formula GL = GI × net carbs ÷ 100 [11]. Both values are shown only when available with reasonable confidence.

5. What the score does not do

• It is not a medical recommendation. It does not assess suitability for any specific medical condition (diabetes, hypertension, kidney disease, food allergy, etc.).
• It is not a substitute for advice from a registered dietitian, physician, or pharmacist.
• It is not a safety or quality test. We do not verify whether a product is actually manufactured to the standards printed on its pack.
• It is not designed for children under 12, pregnant or lactating individuals, or anyone on a clinically prescribed diet. Such users should rely on guidance from their healthcare provider.
• A high score does not mean a product is healthy in unlimited quantities. Portion size still matters.

If you spot a product whose printed label disagrees with what we show in-app, please report it from the product page or email support@tryok.app. We re-extract and re-score on every report.

6. References

1. World Health Organization. Guideline: Sugars intake for adults and children. WHO, 2015. who.int/publications/i/item/9789241549028
2. World Health Organization. Guideline: Sodium intake for adults and children. WHO, 2012. who.int/publications/i/item/9789241504836
3. American Heart Association. Saturated Fat. heart.org — Saturated Fats
4. National Institute of Nutrition (ICMR). Dietary Guidelines for Indians, 2024. nin.res.in — Dietary Guidelines for Indians
5. Jäger R, Kerksick CM, Campbell BI, et al. International Society of Sports Nutrition Position Stand: Protein and Exercise. J Int Soc Sports Nutr. 2017;14:20. doi.org/10.1186/s12970-017-0177-8
6. Food Safety and Standards Authority of India. Food Safety and Standards (Labelling and Display) Regulations, 2020. fssai.gov.in — Labelling & Display Regulations
7. Monteiro CA, Cannon G, Levy RB, et al. Ultra-processed foods: what they are and how to identify them. Public Health Nutr. 2019;22(5):936-941. doi.org/10.1017/S1368980018003762
8. Mifflin MD, St Jeor ST, Hill LA, Scott BJ, Daugherty SA, Koh YO. A new predictive equation for resting energy expenditure in healthy individuals. Am J Clin Nutr. 1990;51(2):241-247. doi.org/10.1093/ajcn/51.2.241
9. Frankenfield D, Roth-Yousey L, Compher C. Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: a systematic review. J Am Diet Assoc. 2005;105(5):775-789. doi.org/10.1016/j.jada.2005.02.005
10. National Heart, Lung, and Blood Institute. Aim for a Healthy Weight: Guide to Behavior Change. nhlbi.nih.gov — Aim for a Healthy Weight
11. Atkinson FS, Foster-Powell K, Brand-Miller JC. International tables of glycemic index and glycemic load values: 2008. Diabetes Care. 2008;31(12):2281-2283. doi.org/10.2337/dc08-1239

7. Disclaimers

The Health Score, daily calorie targets, macronutrient targets, allergen warnings, and any other health-related information presented in the OK app are for general informational and educational purposes only. They are not a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of a qualified healthcare professional with any questions you may have regarding a medical condition or before making any dietary changes. Never disregard professional medical advice or delay seeking it because of something you have read in the OK app.

The scores are computed from information printed on product packaging and are only as accurate as that source. OK makes no warranty as to the completeness or accuracy of any individual product score and disclaims liability for any decision made in reliance on it.

If you experience any adverse reaction after consuming a product, stop consumption immediately and seek medical attention. To report a suspected mislabelled or unsafe product, email info@tryok.app.