A reliable idli dosa batter recipe comes down to six numbers: about 3 parts rice to 1 part urad dal by weight, at least 4 hours of soaking, a smooth grind with cool water, 2% salt by weight, a warm spot near 30 to 32°C, and 8 to 12 hours of fermentation. The batter is ready when it has grown about one and a half times, smells pleasantly sour and is full of small bubbles. Every one of those numbers comes from lab studies of idli fermentation, and the steps below show why each one matters.
- 3 : 1 rice to urad dal by weight, the ratio with the most B vitamins in a 2011 lab study.
- 4 hours minimum soak, rice and urad separately.
- 2% salt by batter weight, about 20 g per kilo.
- 30 to 32°C for 8 to 12 hours is the fermenting window.
- About 1.5 times the starting volume, plus a clean sour smell, means it is ready.
This is the practical companion to my deep dive into idli fermentation science, which covers which bacteria do the work. Here the job is simpler: a method by weight that works in a home kitchen, a calculator for your batch, and a way to read the batter so you know when it is done. If your batter refuses to rise in winter, the dedicated fix is in why idli batter won't ferment in winter.
What ratio of rice to urad dal should you use?
Use 3 parts rice to 1 part urad dal by weight for idli; anything from 2:1 to 4:1 works. Ghosh and Chattopadhyay (2011) fermented 2:1, 3:1 and 4:1 batters with 2% salt and found the 3:1 blend produced the most riboflavin (0.76 mg per 100 g) and thiamine (0.73 mg per 100 g). A 2014 sensory study found the best-liked idli at 3 parts rice to about 1.5 parts dehusked urad, roughly 2:1, fermented for 10.2 hours.
More urad gives softer, fluffier idli because urad holds the gas; more rice gives a firmer idli and a crisper dosa.
In the mid-2000s P. C. Musthafa, back in Bengaluru, noticed something on the shelf of a cousin's grocery shop: packets of idli batter from a local supplier, sold with no brand and no control over quality or hygiene. The cousin tried to get the supplier to fix it and failed, so the family decided to make batter themselves.
They started in a 50 sq ft room, with Musthafa putting in his own savings and a first target of 100 packs a day. It took about nine months to work, then a bigger kitchen, and in 2008 a proper factory. By 2015-16 iD Fresh Food had a turnover of ₹100 crore, and its centres were making enough batter every day for up to 10 lakh idlis (the article calls that an estimate). A factory has to pin down the same numbers a home cook guesses at: ratio, salt, warmth and time.
The Better India: How PC Musthafa built iD Fresh (Feb 2017)| Ratio (rice:urad, by weight) | What you get | Best for |
|---|---|---|
| 2:1 | Softest, highest rise, most protein | Idli; top sensory score in the 2014 study |
| 3:1 | Balanced; peak B vitamins in the 2011 study | One batter for both idli and dosa |
| 4:1 | Firmer idli, crisper and thinner dosa | Dosa-first households |
Weigh, don't use cups. Rice and dal have different densities, so a 3:1 ratio in cups is not 3:1 in weight. Protein is part of the reason the ratio matters: IFCT 2017 gives urad dal 23.1 g protein per 100 g against 7.8 g for parboiled rice, so every extra part of urad lifts the protein of the finished idli.
How long should you soak rice and urad dal?
Soak for at least 4 hours, the time used in the 2011 lab study at 30°C, and up to 6 to 8 hours in a cool kitchen. Soak rice and urad separately, because they grind differently. Rinse them, but don't scrub the grains with hot water or anything antimicrobial: a 2019 study found that grains whose surfaces were sterilised did not ferment at all, because the bacteria that raise idli come in on the grain itself.
Parboiled rice and dehusked urad were soaked 4 hours at 30°C, ground, blended at 2:1, 3:1 and 4:1 with 2% salt, and fermented for up to 10 hours or more while the team tracked viscosity, density, pH, acidity and vitamins. Batter density fell from 0.93 to 0.59 g/cm³ as gas built up, pH dropped from 5.9 to 4.21, and acidity rose from 0.44% to 0.91%. Shear stress peaked at 7 hours. The 3:1 blend made the most riboflavin (0.76 mg/100 g) and thiamine (0.73 mg/100 g), and folic acid peaked at 0.75 mg/100 g after 10 hours. Limit: one lab, one rice and dal variety, controlled temperature, so home timings will differ.
Read the paper (free)How to make idli and dosa batter, step by step
- Weigh. For about 1.1 kg of batter, weigh 300 g idli rice (parboiled) and 100 g whole or split dehusked urad dal. Add a teaspoon of methi (fenugreek) seeds to the urad if you like the traditional flavour and colour on dosa.
- Rinse and soak separately. Rinse twice in cool water. Soak rice and urad in separate bowls with water well above the grain, 4 to 8 hours.
- Grind the urad first. Drain and grind urad with cold water added a little at a time until it is smooth, light and fluffy; this is the gas-holding part of the batter. One well-tested home method adds water to the soaked urad until the total on the scale reads 4 times the dry urad weight (100 g dry urad, 400 g total).
- Grind the rice slightly coarse. Grind the rice to a texture about like fine cornmeal, not perfectly smooth, adding water to the soaked rice until the total on the scale reads 2.25 times the dry rice weight (300 g dry rice, 675 g total). Coarser rice gives idli texture; very fine rice suits dosa.
- Mix by hand and salt. Combine both batters in a container at least twice their volume. Add salt at 1 to 2% of the batter's weight (11 to 21 g for 1.1 kg); the 2011 study used 2%, the Fresh Loaf method about 1%. Mix with your hand for a minute; many cooks find this helps the ferment start.
- Ferment warm. Cover loosely and keep near 30 to 32°C for 8 to 12 hours. In winter, an oven with only the light on or a box with a warm water bottle helps; check with a thermometer.
- Read the batter. It is ready when it has risen about one and a half times, has a domed top with small bubbles throughout, and smells pleasantly sour, not alcoholic or cheesy.
- Stir gently and use, or chill. Stir lightly before steaming idli; thin with a little water for dosa. Refrigerate the rest at once.
Wet grinder or mixie: does it matter?
Both work; the difference is heat and air.
- Stone wet grinder: runs slowly and folds air into the urad, which is why idli shops use one.
- Mixer jar (mixie): spins fast and can warm the batter during long runs, so grind in short bursts and use chilled water.
For the urad, the kitchen test is the same either way: a spoonful dropped into water should float, a sign it is holding air.
How much salt goes in idli batter?
Use 2% of the batter's weight: 20 g per kilo. That is the level used in the 2011 study and in the dhokla fermentation study, which found 8 g per kg optimal for its own batter, so 1 to 2% is the working range. Salt also slows fermentation, so a home cook on The Fresh Loaf forum adds it before fermenting when the weather is hot and after when it is cold; their tests found the type of salt, iodised or not, made little difference.
What temperature does idli batter need?
Around 30 to 32°C. The reference study soaked and fermented at 30°C, and the dhokla optimisation study, which uses a similar rice-and-pulse lactic ferment, found 32°C best. Cooler kitchens slow the ferment, which is why the same recipe rises overnight in summer and sulks in winter. A home cook on The Fresh Loaf forum, who tested this methodically, recommends 30 to 40°C and controls it at 37°C for consistent overnight results; warmer and faster leaves less margin before the batter over-sours.
Using 16S rRNA sequencing, qPCR and culture work, the team tracked bacteria through idli fermentation. Lactic acid bacteria made up 86% of the population, with Weissella the key genus; the community was similar across batters from different regions. Grains whose surfaces were sterilised did not ferment, showing the starter comes from the grain itself. Radiolabelled glucose confirmed that the gas comes from the heterofermentative pathway, and the microbes also broke down starch and phytate and produced B vitamins. Limit: lab batters; household practice (hand mixing, reused containers) was not tested.
Read the paper (free)How do you know the batter is ready?
Look for three signs together:
- Rise: about 1.5 to 1.6 times the starting volume.
- Texture: mousse-like and full of fine bubbles.
- Smell: clean and sour, not alcoholic or cheesy.
The rise is measurable: the 2011 study recorded batter density falling from 0.93 to 0.59 g/cm³, which is 1.58 times the volume. A sharp alcohol or acetone smell or a grey watery layer means it went too far.
Plan your batch
Use this to check what you just read. Enter the rice you will use, the ratio and the water you add while grinding (weigh the jug before and after). The calculator gives the urad and salt to weigh, and the size of container you need so the batter does not overflow, using the densities measured in the 2011 study.
Idli and dosa batter batch planner
How long does idli batter last?
Use fermented batter within 2 to 3 days in the fridge. A 2016 study on idli batter found plain batter kept for 2 days at 30°C, and that adding 5% curry leaf powder stretched it to 5 days by slowing spoilage microbes; refrigeration slows everything further. Keep it covered, use a clean spoon each time, and throw it out if it smells off or grows mould.
What does a plate of idli give you?
The Indian Nutrient Databank's idli recipe works out to 138 kcal, 4.6 g protein, 28.2 g carbohydrate and 0.3 g fat per 100 g, one of the leanest Indian breakfasts. A plain dosa carries much more because of the oil on the tawa and its lower water content. Fermentation itself adds B vitamins and breaks down phytate, which helps the iron and zinc in urad become more available.
| Item (INDB recipe) | Energy (kcal/100 g) | Protein (g) | Carbohydrate (g) | Fat (g) |
|---|---|---|---|---|
| Idli | 138 | 4.6 | 28.2 | 0.3 |
| Masala dosa | 165 | 3.3 | 19.6 | 7.8 |
| Rava idli (semolina) | 206 | 9.3 | 33.7 | 3.5 |
| Uttapam | 256 | 6.2 | 36.3 | 9.0 |
| Plain dosa | 381 | 10.3 | 64.1 | 8.4 |
Should you add poha, cooked rice or baking soda?
Poha or a handful of cooked rice, ground in with the rice, is a common trick for softer idli; it adds soft starch, and the result is a matter of taste. Baking soda gives instant lift but makes the batter alkaline and can leave a soapy taste; it is a shortcut, not a fix.
If your ferment fails, the cause is usually temperature, old urad dal or too much salt in the cold, not missing additives.The troubleshooting rule of this guide
Restaurants have their own tricks: the history of Udupi kitchens, where this batter was scaled up, is in the story of Udupi restaurants. And for the opposite extreme, see what if idli batter fermented for a year.
What's new
- Karnataka bans plastic sheets for steaming idli after inspectors find 52 eateries using them The Week, Feb 2025
- FSSAI asks Karnataka to report on plastic use in idli preparation FoodTechBiz, Mar 2025
Questions people ask
Why didn't my idli batter rise?
Most often the kitchen was too cold, the urad dal was old, or the urad was under-ground and holds no air. Move the batter somewhere near 30°C, give it more hours, and next time grind the urad until a spoonful floats in water.
Can I use the same batter for idli and dosa?
Yes. A 3:1 batter does both; thin a portion with a little water for dosa. Dosa also tolerates older, more sour batter better than idli does.
Should I add salt before or after fermenting?
Before, at about 2%, in a warm kitchen; that is what the lab studies did. In a cold kitchen, add it after fermenting so it does not slow a weak start.
Why does my batter smell of alcohol?
It fermented too long or too warm. Use it for dosa or uttapam quickly, and shorten the time next batch.
Is parboiled rice necessary for idli?
No, but it is the classic choice and the one used in the reference study.
Sources
- Ghosh D., Chattopadhyay P.: Preparation of idli batter, its properties and nutritional improvement during fermentation, J Food Sci Technol (2011; ratios, 4 h soak at 30°C, 2% salt, density, pH, vitamins)
- Mandhania M.H. et al.: Diversity and succession of microbiota during fermentation of idli, Appl Environ Microbiol (2019)
- Durgadevi M., Shetty P.H.: Effect of ingredients on sensory profile of idli, J Food Sci Technol (2014; 3:1.475 ratio, 10.2 h)
- Chelliah R. et al.: Bio-fortification and shelf-life extension of idli batter using curry leaves, J Food Sci Technol (2016; shelf life 2 to 5 days at 30°C)
- Sharma A. et al.: Preparation of antinutrients-reduced dhokla using response surface optimisation, J Food Sci Technol (2018; 8 g/kg salt, 32°C)
- Mukherjee S.K. et al.: Role of Leuconostoc mesenteroides in leavening the batter of idli, Applied Microbiology (1965)
- ICMR-NIN: Indian Food Composition Tables 2017 (urad and rice protein)
- Indian Nutrient Databank (Vijayakumar et al. 2024), recipe values for idli, dosa, uttapam
- The Fresh Loaf forum: Idli, reposted from my blog
- The Better India: PC Musthafa and iD Fresh Food (Feb 2017; source for the story)
- The Week: How safe is your idli? Karnataka cracks down (Feb 2025)
- FoodTechBiz: FSSAI addresses concerns regarding plastic use in idli preparation (Mar 2025)

