Open a mainstream nutrition-tracking app and try to log an ordinary South Indian breakfast: two idlis, a ladle of sambar, a spoon of coconut chutney. Often the app stalls. You get a generic "idli" entry if you are lucky, with nothing that reflects sambar's mix of dal and vegetables, no way to tell coconut chutney from tomato chutney, and no allowance for the ghee a home cook adds that a hostel mess doesn't. Log a full meal of rice, sambar, rasam, a poriyal and a small bowl of curd, and the gap compounds: five separate dishes, none of them barcoded, none of them matching a database built around packaged, labelled food.
This isn't a niche edge case. Rice with sambar and rasam is an everyday plate across Karnataka, Tamil Nadu, Andhra Pradesh, Telangana and Kerala, and it is a plate many tracking apps were simply not designed around.
A database problem, not a discipline problem
Many food-tracking apps are barcode-first: scan a label, read the macros off it, log. That works well for a granola bar. It breaks down for home-cooked or mess-cooked food that never had a label. South Indian cooking is a particularly hard case because one "sambar" varies by household, region and cook, and it is made of several cooked components (dal, tamarind, vegetables, tempering oil) that a single generic database line can't represent.
India's own reference data doesn't solve this on its own either. The Indian Food Composition Tables 2017 from ICMR's National Institute of Nutrition cover 528 key foods, and, except for eggs, the values are for foods in raw form. Cooked dishes like sambar or a batter-based idli have to be worked out from ingredients and recipes.
The practical risk is under-counting, and it tends to hit the nutrient people most want to track: protein. The dal in sambar, the curd, and any egg or paneer on the plate are exactly the parts that get lost when an app has no proper entry to match them against. Someone who eats this way every day can end up with a tracker that quietly misreads their own diet.
There is a framing problem too. A lot of Indian-diet content defaults to a North Indian plate (roti, sabzi, dal, katori portions) as the generic "Indian meal", while people eating rice, sambar, rasam and poriyal daily get comparatively little direct attention.
How to log a South Indian plate more accurately
Whatever app you use, a few habits close most of the gap:
- Log components, not "South Indian meal". Idli, sambar and chutney are three entries. Our breakdown of idli sambar calories for the full plate shows how much the sides add.
- Name the dosa. Plain, masala and rava dosa differ meaningfully; see dosa calories compared.
- Count the ladle, not just the rice. In a rice-based meal, rice portion and added fat move the total most. Our guide to a South Indian meal plate for weight loss covers proportions.
- Don't drop the curd. It is easy to forget and it carries protein. Curd rice is a meal in its own right, not a garnish.
All dish values are approximate: recipes and portions vary by kitchen.
Building the data dish by dish
NYUS, the India-first AI fitness and nutrition coach I build, is currently in early access. Its core bet is that this has to be fixed by mapping food dish by dish rather than adding Indian names to a Western database. The app has 9,000+ Indian and global dishes with macros, including idli, dosa, sambar and rice-based South Indian plates as first-class entries, and you can log by photo, barcode or search, in English, Hindi or Hinglish.
That doesn't make every kitchen's sambar identical, and I'm not claiming NYUS has solved South Indian food logging with perfect precision. Recipes vary by household and no database captures every cook's proportions. What it changes is the starting point: a real entry instead of a guess or "no match found".
The other core mechanic is weekly rather than daily recalibration. The coach reads what you actually logged in training, nutrition and recovery over the week and adjusts the plan, rather than resetting a streak the moment one day doesn't fit. It reads steps, sleep and heart data from compatible wearables through Health Connect, and it has no ads and doesn't sell user data. It is 30 days free with no card, then ₹499 a month, with no auto-renewal.
Why the gap deserves attention
This is a daily experience for anyone in South India who has tried to track their diet seriously in a mainstream app and hit a wall the moment breakfast wasn't a cereal box. It isn't fully solved, by NYUS or anyone else. NYUS is early access and still finding its own gaps. But the underlying critique, that nutrition-tracking tech has a real blind spot for South Indian home food, is worth more scrutiny than it has had.
This article describes app features and general nutrition information; it is not medical or dietary advice. Consult a qualified professional for individual guidance.