Dev Narsinghani: writing

How to Read a Food Label in 10 Seconds

2026-09-30T09:00:00+05:30

I was Chief of Staff at Food Pharmer and on the founding team at Only What's Needed. Reading the back of food packs was a big part of that work.

This is the fast version I use in a shop, basket in hand.

You can skip the nutrition course. It takes 5 checks, and each one takes about 2 seconds.

Here is the exact order I use.

Step 1, skip the front (0 seconds)

The front is marketing. It's where the brand picks the words, the photos and the claims.

The back is different. The law decides what goes there, and the brand can't skip it.

So I turn it over before I read a single word up there. Words like natural, wholesome and multigrain tell me nothing about what's inside.

Train your hand to flip first, because everything below happens on the back.

Step 2, read the top 3 ingredients (2 seconds)

In India, the ingredient list must run from most to least by weight. That's an FSSAI rule, and it's the most useful rule on the pack.

It means the first 3 names are most of what you're eating.

Take a mixed fruit jam. The one I checked puts sugar 1st and the fruit 2nd.

The front of the jar shows fruit. The back puts sugar ahead of the fruit.

The masala instant noodle pack I checked lists maida first and palm oil next.

Nothing wrong with eating that now and then. Just know the block is mostly refined flour.

My rule is simple. If sugar, maida or oil sits in the top 3, it's a treat for me.

Step 3, count every name for sugar (2 seconds)

Here's the trick that fools smart people. A brand can split sugar across several names.

Each name then sits lower on the list, so each one looks small.

Names I watch for on Indian packs:

sugar
invert sugar syrup or invert syrup
liquid glucose or glucose syrup
dextrose
fructose
honey
jaggery
fruit juice concentrate

The WHO counts honey, syrups and juice concentrates as free sugars, the kind to limit. So I count them too.

Honey cornflakes are a good example. The pack I checked lists sugar, dextrose and honey.

That's 3 sweeteners in one bowl.

The glucose biscuit I checked does the same, with sugar, invert syrup and liquid glucose.

If I count 2 or more sugar names, I stop reading and put it back.

Step 4, read sugar per 100g (2 seconds)

Every pack must show its numbers per 100g, or per 100ml on a drink. Brands also show a per serving column, and they pick the serving size.

A tiny serving makes every number look small. Per 100g is the same for every pack, so it's the only column I read.

2 lines help me judge the sugar number fast:

5g or less per 100g is low. That's FSSAI's legal line for a low sugar claim, about a teaspoon.
More than 22.5g per 100g is high. That's the UK's red-light line, close to a quarter of the pack.

Now the real labels:

The mixed fruit jam I checked: about 68g sugar per 100g, so two-thirds of the jar is sugar.
The honey cornflakes I checked: 25g per 100g, so a quarter of every bowl is sugar.
The glucose biscuit I checked: about 28g per 100g, so more than a quarter of every biscuit is sugar.

For drinks, FSSAI's low sugar line is 2.5g per 100ml. That's about half a teaspoon in every 100ml.

Step 5, turn sodium into salt (2 seconds)

This is the check almost nobody does, and it's the one I use most.

Indian labels list sodium, sometimes in mg and sometimes in g. 970mg is 0.97g, just under 1g. The daily limit is for salt, and salt is roughly 2.5 times the sodium.

So the quick sum is sodium x 2.5 = salt.

The WHO says adults should eat less than 5g of salt a day. That's just under a teaspoon.

Back to the masala instant noodles:

The pack I checked shows 0.97g sodium per 100g.
The block I checked is 70g, so it carries about 0.68g of sodium.
Times 2.5, that's about 1.7g of salt.

One block of noodles is about a third of your whole day's salt. That's before the dal, the pickle or the namkeen.

2 lines help here too.

0.12g sodium or less per 100g is low. That's FSSAI's legal line for a low sodium claim.
More than 0.6g sodium per 100g is high. That's the UK red-light line, 1.5g of salt or about a quarter teaspoon.

Salty food rarely tastes salty enough to warn you. The label does it for you, if you do the sum.

What a good label looks like

This isn't about fear. Plenty of packs pass.

The peanut butter I buy lists one thing, roasted peanuts. The Greek yogurt I buy lists just milk, milk solids and cultures.

A short list with real food at the top is the best sign on a pack.

The 10-second checklist

Save this for your next grocery run:

Turn it over and skip the front
Read the top 3 ingredients
Count the sugar names (2 or more is a warning)
Sugar per 100g: 5g (about a spoon) is low, over 22.5g is high
Sodium x 2.5 = salt, and over 0.6g sodium per 100g is high

Do it on 5 packs in your own kitchen tonight. I bet one of them surprises you.

Which one did it for you, the sugar or the salt?

Sources

Every number above comes from one of these.

FSSAI, Food Safety and Standards (Labelling and Display) Regulations, 2020, Reg 5(2)(b) (ingredients in descending order by weight) and Reg 5(3) (nutrition information per 100g or 100ml, including total and added sugars)
FSSAI, Food Safety and Standards (Advertising and Claims) Regulations, 2018, Schedule I (low sugar: not more than 5g per 100g or 2.5g per 100ml; sugar free: not more than 0.5g; low sodium: not more than 0.12g per 100g)
WHO, Guideline: Sugars intake for adults and children, 2015 (free sugars include honey, syrups and fruit juice concentrates)
WHO, Sodium reduction fact sheet (adults: less than 5g salt a day, about 2g sodium; salt is about 2.5 times sodium)
UK Department of Health and Food Standards Agency, Guide to creating a front of pack nutrition label for pre-packed products, 2016 (high: sugar over 22.5g per 100g, salt over 1.5g per 100g)
Open Food Facts label transcriptions of 4 packs sold in India, one per category, read 28 September 2026: a mixed fruit jam (sugars 68.1g per 100g), honey cornflakes (sugars 25g per 100g), a glucose biscuit (sugars 27.7g per 100g) and masala instant noodles (sodium 0.97g per 100g, 70g block)
Peanut butter and Greek yogurt lists: pack labels as shown on Zepto product pages, read 8 September 2026

Why I Sleep 8 Hours Every Night

2026-09-30T09:00:00+05:30

I was Chief of Staff at Food Pharmer and on the founding team at Only What's Needed. My work has been about what goes on our plates.

The habit I protect hardest has nothing to do with food. I sleep from 11pm to 7am, a full 8 hours, and I've done it since school.

Around me, 5 or 6 hours is normal. People say it with a kind of pride, like it proves how hard they work.

Those 8 hours are the smartest thing I do for my work and body.

Here's why, and here's what the research actually says.

How much sleep an adult needs

The American Academy of Sleep Medicine and the Sleep Research Society looked at the evidence together. Their joint advice is plain.

Adults should sleep 7 hours or more a night, on a regular basis.

Their report links regularly sleeping less to weight gain, diabetes and high blood pressure. Heart disease, depression, mistakes and accidents are on that list too.

7 hours is the floor in that advice. I aim for 8 because I want room for a bad night.

Short sleep makes you worse at your job, and you can't feel it

This is the study that changed how I think about sleep.

In 2003, a lab kept healthy adults on a fixed amount of sleep for 2 weeks. One group got 6 hours in bed a night.

By the end, that group did as badly on tests as people kept awake for a full night. Their attention and working memory slipped almost every day.

Their own sleepiness ratings barely rose after the first few days.

They felt only a little tired, while their scores kept falling.

A group on 4 hours did worse still, matching 2 full nights awake.

So when someone tells me they're fine on 5 hours, I believe they feel fine. The research says feeling fine is a poor guide.

Short sleep makes you hungrier

In a 2004 study, young men slept only 4 hours a night for 2 nights. The same men also had 2 nights of 10 hours, for comparison.

After the short nights, their hunger went up by 24%, almost a quarter. The hormone that signals fullness dropped, and the one that drives hunger rose.

Their cravings rose most for sweet, salty and starchy food.

I've read food labels for work, so this one matters to me. When you are tired, you reach for exactly the packs I warn people about.

Short sleep may change what a diet takes off your body

This study looked at what happens to a diet when sleep is cut short.

In 2010, 10 overweight adults ate a reduced diet for 2 weeks, twice. One round they had 8.5 hours in bed each night, and one round 5.5 hours.

They lost about the same weight both times, but what they lost was very different.

On short sleep, the fat they lost fell by 55%, so less than half as much.
The loss of lean tissue, like muscle, went up by 60%.

Both rounds had the same diet and showed the same drop on the scale. The difference was in what that weight was made of.

It's a small study, and it only ran 2 weeks. Treat it as a signal, not proof.

Keeping the same hours may matter as much as the total

A 2024 study tracked sleep for about 60,000 people in the UK with wrist devices.

In that study, more regular sleep was linked to a lower risk of dying over the follow-up. Keeping the same hours predicted that risk better than total hours did.

The authors suggest a simple target: fall asleep and wake up within about a 1-hour window every day.

This is the part of my habit I'd defend most. 11 to 7 is the same on a Monday and a Saturday, give or take.

Some people really can live on less

I want to be fair here. A few people carry rare gene changes that let them do fine on about 6 hours.

The first one was found in 2009, in a single family. It's rare, and most of what we know comes from a handful of families.

If you need 2 coffees to get going, you're probably not one of them.

What I actually do

3 habits keep my 11 to 7 steady:

The same bedtime and wake time, weekends included
Daylight on my face soon after I wake
A dark bedroom, with the curtains fully shut

The research backs all 3.

The 2024 study's authors suggest keeping sleep and wake times inside about a 1-hour window.
Light is the body clock's main signal, and experts say daylight is the best daytime source.
The same experts say the bedroom should be as dark as possible while you sleep.

One more habit is worth trying, and it's about the lit screen before bed.

In one study, a lit tablet before bed delayed the sleep hormone by over 1.5 hours. That study compared a lit tablet with a printed book, in the same dim room. Tablet readers took longer to fall asleep and felt less alert the next morning.

One more rule worth stealing

Watch the time of your last coffee.

In a 2013 study, people took a big dose of caffeine 6 hours before bed. The dose was about 2 or 3 regular coffees.

It still cut their sleep by more than 1 hour on a home sleep monitor.

For an 11pm bedtime, 6 hours back is 5pm. That's the latest coffee this study supports.

Why this belongs next to food

Most of my work has been about labels, sugar and salt. Sleep turned out to matter just as much, and it has no label at all.

It costs nothing and it's boring to talk about. Research links it to your hunger, your focus and what a diet takes off your body.

If you try one thing, fix your wake-up time for 2 weeks. Keep it inside a 1-hour window, weekends too.

Are you a 5-hour sleeper or an 8-hour sleeper right now?

Sources

Every number above comes from one of these.

Watson NF et al, Recommended amount of sleep for a healthy adult, joint consensus statement of the American Academy of Sleep Medicine and Sleep Research Society, Sleep 2015, 38(6), 843-844, doi 10.5665/sleep.4716
Van Dongen HPA, Maislin G, Mullington JM, Dinges DF, The cumulative cost of additional wakefulness, Sleep 2003, 26(2), 117-126, doi 10.1093/sleep/26.2.117
Spiegel K, Tasali E, Penev P, Van Cauter E, Sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite, Annals of Internal Medicine 2004, 141(11), 846-850, doi 10.7326/0003-4819-141-11-200412070-00008
Nedeltcheva AV, Kilkus JM, Imperial J, Schoeller DA, Penev PD, Insufficient sleep undermines dietary efforts to reduce adiposity, Annals of Internal Medicine 2010, 153(7), 435-441, doi 10.7326/0003-4819-153-7-201010050-00006
Windred DP et al, Sleep regularity is a stronger predictor of mortality risk than sleep duration, Sleep 2024, 47(1), zsad253, doi 10.1093/sleep/zsad253
He Y et al, The transcriptional repressor DEC2 regulates sleep length in mammals, Science 2009, 325(5942), 866-870, doi 10.1126/science.1174443
Brown TM et al, Recommendations for daytime, evening, and nighttime indoor light exposure to best support physiology, sleep, and wakefulness in healthy adults, PLOS Biology 2022, 20(3), e3001571, doi 10.1371/journal.pbio.3001571
Chang AM, Aeschbach D, Duffy JF, Czeisler CA, Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness, PNAS 2015, 112(4), 1232-1237, doi 10.1073/pnas.1418490112
Drake C, Roehrs T, Shambroom J, Roth T, Caffeine effects on sleep taken 0, 3, or 6 hours before going to bed, Journal of Clinical Sleep Medicine 2013, 9(11), 1195-1200, doi 10.5664/jcsm.3170

The AI Marketing Stack I Built Without Being an Engineer

2026-09-30T09:00:00+05:30

I was Chief of Staff at Food Pharmer and on the founding team at Only What's Needed. My work was food labels and brand, and I'm not an engineer.

Over the last few months I built a small AI system to help with my own LinkedIn. It drafts, checks, designs and learns.

The writing and image models run free on my own laptop.

Here is what's inside, and the lesson each part taught me.

Part 1, a study before any tool

I didn't start with a tool. I started by studying what already worked for 9 health creators.

Each post was scored against its own creator's usual numbers. A big account can't win just by being big.

That study became my guide. 2 findings changed what I make:

Posts that turned a number into something you can picture did 2.3 times better. Half the jar is sugar is easier to feel than a gram count.
Carousels did about 1.7 times better than the creator's normal post.

The lesson for marketers is simple. Compare each post with the same creator's normal, never with the whole feed.

Part 2, rules that live in code

I have a lot of writing rules. No em dashes, no sentence starting with And, and a one-line hook.

For a while those rules lived in a document. I kept breaking them, and so did the AI.

So I moved them into 2 checkers. One runs while I draft and scores the post. The other runs on the final draft and flags anything that breaks a rule.

Some rules it enforces:

The first line must be one sentence and fit on a phone screen.
Words a 10-year-old wouldn't know get flagged, each with a plain swap.
Brand names in a negative frame get flagged.
A number with no source gets flagged.

This lesson is the most useful one in the whole setup, so here it is plainly. A rule you have to remember is one you'll break. Write it as a check.

Part 3, a learning loop that can't overreact

I note the likes, comments and reposts on my own posts. A script turns that into weights.

Those weights tell the writer which topics, formats and visuals are working for my audience.

I put 3 limits on it:

Nothing moves off neutral until there are at least 4 posts of data.
The top 10% of results are capped, so one viral post can't dominate.
Every weight stays between half and 1.6 times normal.

One lucky post should never rewrite your whole strategy. The lesson is to make your learning loop slow on purpose.

Part 4, visuals where AI never draws a number

AI image models are great at mood and bad at exact digits. A wrong digit on a health post would hurt trust in me more than anything else.

So every number and table is built as a web page, then screenshotted with Playwright. That gives a sharp image every time.

For photo-style scenes I run FLUX.1-dev locally through ComfyUI. It makes the background, and the real numbers get laid on top.

The lesson is to split the job. Let the image model make the picture, and never let it write a number.

Part 5, small local models, each with one job

I use Ollama to run open models on my own graphics card. Each model does a single task:

A drafting model writes a first version from a fact sheet. I picked it in a blind test of 5 models.
A second model reads each draft the way a 10-year-old would, and flags anything confusing.
A vision model reads a photo of a pack and turns the label into a fact sheet.

The label reader reads every photo twice. If the 2 reads disagree on a number, it's thrown out. A misread digit can't slip through.

I still read every draft, and I bin the ones that miss. I have thrown out more than a few.

The lesson is to test models blind on your own task. The famous name isn't always the one that writes best for you.

Part 6, a second AI whose job is to disagree

This is the part I'm proudest of.

I had a batch of sleep posts with research behind each claim. First, every claim went to one AI agent to check against the original paper.

A second agent then tried to prove the first one wrong.

23 of the 24 fact sheets needed a fix, from wrong numbers to wrong papers. A few said more than the study did.

None of those errors went live, because the check ran first.

The lesson for any marketer using AI is blunt. AI sounds sure of itself even when it's wrong. Make checking cheap, and have a second AI try to prove the first one wrong.

What you can steal without writing code

You don't need my setup to use these ideas. Here's the short version:

Study your own niche's best posts before you copy a format.
Judge each post against the same creator's normal.
Turn your style guide into a checklist something can run.
Keep your AI images away from facts and numbers.
Test AI tools blind on your real work.
Have a second pass that tries to prove the first one wrong.

I built all of this with AI tools after hours, one small piece at a time. The hard part was knowing the problem well, and the code came second.

Which one would help your team first, the checker or the second AI that disagrees?

Sources

Every number above comes from one of these.

Author's own review of public posts from 9 health and food creators, 2026, each compared with that creator's usual engagement
Author's own citation audit of 24 sleep fact sheets, August 2026 (one checking agent per claim, then one refuting agent; 23 needed corrections)
Playwright, open-source tool by Microsoft that turns a web page into an image, playwright.dev
FLUX.1-dev image model by Black Forest Labs; ComfyUI, open-source image workflow tool, comfy.org
Ollama, local model runner, ollama.com

How to Read an Indian Food Label: A Plain Guide

2026-09-30T09:00:00+05:30

The front of a pack is mostly the brand's photos, big words and claims. The law decides what every label must show, like the food's real name, the veg mark and the nutrition table [1].

This page covers what an Indian label must show, and how to read it. Every number links to its source.

Short on time? Read Read a food label in 10 seconds. Have a pack in hand? Paste its ingredients into the Label Checker.

What must every Indian food label show?

Packed food in India follows rules from FSSAI, India's food safety regulator: the Labelling and Display Regulations, 2020 [1]. Most packed food must show the food's real name on the front, an ingredient list and a nutrition table. It also needs a veg or non-veg mark, the FSSAI logo and licence number, a batch number and the dates. Allergens must be listed on their own [1].

Name. The name on the front must say what the food really is. A brand name can sit next to it [1].
Ingredients. Listed from most to least by weight, as the food was made. A single-ingredient food can skip the list [1].
Nutrition table. Per 100 g or 100 ml, or per pack for a single-serve pack. It must list energy, protein, carbohydrate and fat. It must also list total sugars, added sugars, saturated fat, trans fat and sodium in mg. Saturated and trans fat can be left out when total fat is 0.5% or less. Cholesterol is needed only when the food has animal fat and more than 0.5% fat. The table also shows the serving size and the number of servings in the pack [1]. Saturated fat and trans fat are the kinds of fat that ICMR-NIN, India's national nutrition institute, asks people to limit [5]. A few simple foods can skip the table, like salt, plain spices, plain tea and coffee, and unprocessed single-ingredient foods [1].
Veg or non-veg mark. Veg is a green filled circle inside a square with a green outline. Non-veg is a brown filled triangle inside a square with a brown outline [1]. Packaged water, soda water, liquid milk, milk powder and honey do not need it [1].
What counts as non-veg. Any food with animal parts, eggs or products of animal origin. Milk, milk products and honey do not count [1].
FSSAI logo and licence number. It must stand out from the background. The brand owner's licence number is a must. If someone else made or packed it, their licence number is needed too [1].
Dates. The date of manufacture or packing and an 'Expiry' or 'Use by' date are a must for most foods. A few, like solid sugar and hard boiled sweets, can skip the expiry date. 'Best before' is optional [1]. Food past its best before date may still be safe, though its quality may have dropped [1].
Batch number. A batch, lot or code number is a must [1].
Allergens. Listed separately after the word 'Contains': cereals with gluten (wheat, rye, barley, oats, spelt), shellfish like prawns and crabs (written as 'Crustacean'), milk, eggs, fish, peanuts and tree nuts (both written as 'Nut'), soy, and sulphite, a preservative, at 10 mg per kg or more [1].
Net quantity, MRP and customer care. These follow a different law on weights and measures [1].

Why read per 100 g and not per serving?

Per 100 g is the same for every pack, so you can compare two packs fairly. The brand picks the serving size, and a tiny serving makes every number look small. A pack can also hold more than one serving. Check the servings in the pack, then scale the numbers to what you actually eat [5].

The law makes packs show nutrition per 100 g or 100 ml, and a per-serve % of RDA [1]. Use the first to compare packs. Use the second only after you check the serving size.

What does % RDA on the pack mean?

RDA is short for Recommended Dietary Allowance. % RDA is the share of a day's amount that one serving gives you. FSSAI works it out from fixed daily figures: 2000 kcal, 67 g total fat, 22 g saturated fat, 2 g trans fat, 50 g added sugar and 2000 mg sodium (5 g salt) [1].

The 50 g for added sugar is well above what ICMR-NIN tells adults to stay within: 20 to 25 g a day in its main advice, and 25 to 30 g in its meal plans [1][5]. So '10% RDA' of added sugar is 5 g, a fifth of 25 g.

How do I find hidden sugar?

Read the ingredient list, not the front. Sugar can be listed under many names, like dextrose, liquid glucose, invert syrup, malt extract, honey and jaggery. A pack can use several of them, so each one sits lower on the list and looks small. Ingredients are listed from most to least by weight [1]. Count every sugar name, and check if one sits in the first 3 ingredients.

In law, 'sugars' means the small sugars made of one or two sugar units, like glucose, fructose, maltose, sucrose and lactose. 'Added sugars' means those sugars added to a food or drink [1].
The nutrition table must show added sugars as well as total sugars [1].
FSSAI lets a pack write 'Sugar' for sucrose, and 'Dextrose' or 'Glucose' for dextrose [1].
ICMR-NIN warns that a 'sugar free' food can still hold fats, refined flour and hidden sugars like maltitol, fructose, corn syrup and molasses [5].
Maltodextrin is made from starch, as chains of glucose. Most of it is not a sugar in FSSAI's legal definition [1], so most of it shows in the carbohydrate line, not the sugar line.

Sugar names to look for

Sugar names on Indian labels, and what each one is
Name on the label	What it is
Sugar, sucrose	Plain white sugar. FSSAI lets a pack write just 'Sugar' for sucrose [1]. Cane sugar is the same sugar. Sugar syrup is sugar melted in water.
Brown sugar, raw cane sugar	Sugar with some molasses left in or added back.
Cane juice, sugarcane juice	Juice pressed from sugarcane, dried or boiled down.
Khandsari, khaand, bura, misri	Traditional Indian sugars. FSSAI has a standard for khandsari, bura and misri [3].
Jaggery, gur, gud, shakkar	Cane or palm juice boiled down into a block or powder [3]. FSSAI has a standard for it [3]. ICMR-NIN lists jaggery as an added sugar [5].
Coconut sugar, palm sugar	Sugar made from palm sap.
Dextrose, glucose	A simple sugar. FSSAI lets a pack write 'Dextrose' or 'Glucose' for it [1].
Trehalose	A sugar made of two glucose units. It counts as a sugar under FSSAI's definition [1].
Liquid glucose, glucose syrup	A sweet syrup. FSSAI's rules for sweets list liquid glucose as a sweetener [4].
Dried glucose syrup, glucose syrup solids	Glucose syrup dried into a powder. FSSAI has a standard for it [3].
Invert sugar, invert syrup, golden syrup	Sugar split into glucose and fructose. FSSAI defines golden syrup as sugar syrup that has been split this way [3].
Fructose, fruit sugar	A simple sugar. ICMR-NIN names fructose as a hidden sugar [5].
Corn syrup, high-fructose corn syrup	Syrups made from corn starch. ICMR-NIN names corn syrup as a hidden sugar [5].
Rice syrup, maple syrup, agave	Sweet syrups. WHO counts the sugars in syrups as free sugars (sugar added to food, plus honey, syrups and juice) [7].
Maltose, malt extract, malt syrup	Maltose counts as a sugar under FSSAI's definition [1]. Malt extract is a sweet syrup or powder made from malted grain.
Lactose	Milk sugar. It counts as a sugar under FSSAI's definition [1], and FSSAI's rules for sweets list it as a sweetener [4].
Honey	WHO counts the sugar in honey as free sugar [7].
Molasses	The dark syrup left over when sugar is made. ICMR-NIN names it as a hidden sugar [5].
Fruit juice concentrate	Juice with the water taken out. WHO counts it as free sugar [7]. A 'no added sugar' food cannot use it in place of sugar [2]. Plain fruit juice also counts as free sugar for WHO [7]. Lemon and lime juice concentrate are used for sourness, so the checker does not count them.
Date paste, date syrup, dried fruit paste	Fruit cooked down into a paste or syrup. A 'no added sugar' food cannot use dried fruit paste in place of sugar [2].
Jam, jelly, sweetened chocolate, sweetened fruit pieces	Sugary ingredients. A 'no added sugar' food cannot contain them [2].
Sweetened condensed milk	Milk boiled down with a lot of sugar added. Unsweetened condensed milk is not counted.
Maltodextrin	Maltodextrin is made from starch, as chains of glucose. Most of it is not a sugar in FSSAI's legal definition [1], so most of it shows in the carbohydrate line, not the sugar line.
Maltitol, sorbitol	Sweeteners made from sugar, with fewer calories. They are not sugars in law, so the tool does not count them. ICMR-NIN says to watch for maltitol in sugar free foods [5].

The Label Checker looks for every name in this table.

How do I turn sodium into salt?

Multiply sodium by 2.5. Salt is 40% sodium, so salt weighs about 2.5 times its sodium [5]. Labels often give sodium in mg, so divide by 1000 first to get grams.

The label says 800 mg sodium per 100 g.
800 mg is 0.8 g.
0.8 x 2.5 = 2 g of salt per 100 g.
That is 40% of the 5 g a day adult limit for salt, in 100 g.

FSSAI and WHO use the same pair: 2000 mg of sodium is 5 g of salt [1][6].

How much salt and sugar is too much in a day?

WHO tells adults to keep salt under 5 g a day in total. That is 2000 mg of sodium, or about one teaspoon [6]. ICMR-NIN gives the same 5 g line for adults [5]. For sugar, ICMR-NIN tells adults to avoid it or keep it to 20 to 25 g a day. Its sample meal plans allow 25 to 30 g [5].

Free sugars are sugars added by the maker, the cook or you, plus the sugar in honey, syrups and fruit juices. The sugar inside fresh fruit and vegetables, and the sugar that is naturally in milk, does not count [7].
WHO says free sugars should stay under 10% of a day's energy. WHO suggests going below 5%, which is about 25 g or 6 teaspoons a day [7].
A lot of it comes from packed food. WHO says one can of sugary soda has up to 40 g of free sugars [7].
ICMR-NIN sets cut-off numbers for packed or cooked food with added sugar, salt or fat. Per 100 g: under 250 kcal, under 3 g added sugar, under 4.2 g added fat and under 0.625 g salt. Per 100 ml of a drink: under 70 kcal, under 2 g added sugar, under 1.5 g added fat and under 0.175 g salt [5]. It calls a food high in fat, sugar or salt when it gives more than 250 kcal per 100 g because of added sugar or fat [5]. These are guidelines, not legal limits. FSSAI has told the Supreme Court its proposed warning limits would be based on these ICMR-NIN guidelines. The exact numbers are not out yet [10].

What do 'no added sugar' and 'natural' mean?

'No added sugar' means no sugar of any kind was added, no sugary ingredient went in, and nothing was used in place of sugar. If the food has its own sugars, the pack must also say 'Contains naturally occurring sugars' [2]. 'Natural' can only describe a single food with nothing added. A food made of several ingredients can at most say 'made from natural ingredients' [2].

No added sugar, in full. No sugar of any kind is added, like sucrose, glucose, honey, molasses or corn syrup. No sugary ingredient goes in, like jam, jelly or sweetened chocolate. Nothing is used in place of sugar, like fruit juice concentrate or dried fruit paste. The making process must not raise the sugar either. For example, the maker cannot use enzymes (helpers that break food down) to turn starch into sugar [2].
Low sugar. Not more than 5 g of sugars per 100 g for solids, or 2.5 g per 100 ml for liquids [2].
Sugar free. Not more than 0.5 g of sugars per 100 g or 100 ml [2].
Natural. Only basic processing is allowed, like roasting, drying, freezing, heating to kill germs (pasteurising) or fermenting. A food additive taken from a natural source by simple physical steps can also be called natural [2]. The same rule covers 'real' and 'genuine' [2].
Pure. Only for a single-ingredient food with nothing added. A mixed food can at most say 'made with pure ingredients' [2]. If 'pure' is in a brand name and could mislead, the pack needs the brand-name disclaimer below [2].
Brand names. Some brand names use words like natural, fresh, pure, original, traditional, authentic, genuine or real. If that could mislead, the front of the pack must print a line. It says: 'This is only a brand name or trademark, or fancy name and does not represent its true nature'. The smallest letter size goes from 1.5 mm on a small pack to 3 mm on a large one [2].
Home-made. Words like 'home-made' and 'home cooked' are not allowed where they could mislead [2].
Disease claims. No food claim can say the food prevents, eases, treats or cures a disease, unless another FSSAI rule allows it [2].

What is the front-of-pack warning FSSAI proposed?

It is a proposal, not a law yet. FSSAI has told the Supreme Court it plans a red hexagon warning on the front of packs, naming each nutrient over the limit: added sugar, added fat or salt [10]. On 28 September 2026, the Court heard the case. It will give its decision later [12]. The final design, limits and dates will come after the judgment.

Status on 30 September 2026. The red hexagon is not a rule yet. FSSAI has not published it as a draft or as a final rule, and no limits are out. The 2022 star-rating draft was never made final. FSSAI told the Court in January 2026 that it plans to withdraw it [9][10][12].
How it would look. A red hexagon on a white square. One nutrient over the limit would be enough to trigger it [10].
Where the limits would come from. ICMR-NIN's Dietary Guidelines for Indians, 2024. They would apply however processed the food is [10]. See the ICMR-NIN figures above.
Sweeteners. A separate warning was also proposed for foods with a sweetener that has no calories, like the ones in diet drinks. The proposed words were 'CONTAIN NON CALORIC SWEETENER' [10].
Timeline FSSAI asked for. About 4 months to prepare draft rules. Then a round of public comments. After the final rules come out, a 365-day period when using the label is a choice, not a must, so companies can use up packs already printed [11][12].
The Court's questions. The judges asked why FSSAI needs a 1-year period when the label is a choice, not a must, and another round of public comments. Both sides had 3 days to send their last points in writing [10][12].

How it got here

13 September 2022. FSSAI's draft star rating for packs, the 'Indian Nutrition Rating' [8][9]. About 14,000 comments came in on it by April 2023 [9].
9 April 2025. The Supreme Court told FSSAI's expert committee to finish its recommendations within 3 months [8].
9 October 2025. A committee of senior government officials said the star draft was not fit to finalise in its current form [9].
January 2026. FSSAI told the Court it plans to withdraw the star draft, because people did not agree on it [9].
February 2026. The Supreme Court asked FSSAI to consider warning labels on the front of packs [11].
10 September 2026. The Court asked FSSAI more questions about its plan to bring the warning in two steps, including the limits and the timeline. FSSAI then proposed one warning for any single nutrient over the limit [10][11].

Checked on 30 September 2026. The Court has not given its judgment yet.

The short version

Flip the pack and read the back.
Read per 100 g, not per serving.
Check the first 3 ingredients.
Count every sugar name.
Turn sodium into salt: times 2.5.

Look for foods you know at the top of the list.

Sources

FSSAI, Food Safety and Standards (Labelling and Display) Regulations, 2020. Compendium Version VIII, 9 September 2025. https://fssai.gov.in/upload/uploadfiles/files/Comp_Labelling%20Display_Version%20VIII_09_09_2025.pdf
FSSAI, Food Safety and Standards (Advertising and Claims) Regulations, 2018. Compendium Version IV, 14 December 2022. https://fssai.gov.in/upload/uploadfiles/files/Compendium_Advertising_Claims_Regulations_14_12_2022.pdf
FSSAI, Food Products Standards, Chapter 2.8: Sweetening agents including honey. https://www.fssai.gov.in/upload/uploadfiles/files/9_%20Chapter%202_8%20(Sweetening%20agents%20including%20Honey).pdf
FSSAI, Food Products Standards, Chapter 2.7: Sweets and confectionery, 1 September 2023. https://fssai.gov.in/upload/uploadfiles/files/Chapter%202_7%20(Sweets%20and%20Confectionary).pdf
ICMR-NIN, Dietary Guidelines for Indians, 2024. https://nin.res.in/dietaryguidelines/pdfjs/locale/DGI_2024.pdf
WHO, Sodium reduction, fact sheet. https://www.who.int/news-room/fact-sheets/detail/sodium-reduction
WHO, WHO calls on countries to reduce sugars intake among adults and children, 4 March 2015. https://www.who.int/news/item/04-03-2015-who-calls-on-countries-to-reduce-sugars-intake-among-adults-and-children
Supreme Court of India, 3S and Our Health Society v Union of India, order of 9 April 2025 (Indian Kanoon). https://indiankanoon.org/doc/160707635/
ThePrint, FSSAI puts brakes on star rating of packaged food, 17 February 2026. https://theprint.in/judiciary/fssai-puts-brakes-on-star-rating-of-packaged-food-underlines-need-for-more-research-before-sc/2857081/
LiveLaw, Supreme Court asks FSSAI why one year is needed for front-of-pack labels, reserves judgment, 28 September 2026. https://www.livelaw.in/top-stories/why-one-year-needed-to-place-front-of-pack-labels-on-packaged-food-supreme-court-asks-fssai-reserves-judgment-552122
LawBeat, Front-of-pack labelling to be rolled out in single phase: FSSAI tells Supreme Court, 26 September 2026. https://lawbeat.in/top-stories/front-of-pack-labelling-to-be-rolled-out-in-single-phase-fssai-tells-supreme-court-1635388
Business Standard, SC reserves order on front-of-pack labels, questions FSSAI over timeline, 28 September 2026. https://www.business-standard.com/amp/india-news/sc-reserves-order-on-front-of-pack-labels-questions-fssai-over-timeline-126092801182_1.html