Homemade cat treats in India: the 22 kcal day, the liver ceiling and what a cat can actually taste

What a cat's treat allowance actually is, in calories

A cat's treat allowance is 18 to 28 kilocalories a day, depending on its weight, and for a 4 kg cat it is 22. The World Small Animal Veterinary Association publishes that figure by body weight and states the rule behind it, that treats should always make up less than 10 percent of a cat's daily calorie intake and should never replace a meal, because every cat needs a complete and balanced diet.

The same guide defines a treat more widely than most owners do. Treats include any food items given in addition to a complete and balanced diet, including snacks, chews, supplements, and foods used to administer medication, whether commercially available or home prepared using food intended for people. The crumb of chicken you dried on Sunday, the paneer you hid a tablet in, and the forty flakes you handed out during a clicker session all come out of one budget. Our guide to giving medication to a pet at home covers the tablet half; the cost of the vehicle you hide it in belongs here.

The allowance column is WSAVA's. The three gram columns are Bonbeno's arithmetic on India's own food composition tables, which measure skinless chicken breast at 704 kilojoules per 100 g, chicken liver at 518 and chicken gizzard at 386, which convert to 1.68, 1.24 and 0.92 kilocalories in every gram. Divide the allowance by those and you get the grams. Nothing here needs a calculator twice.

WSAVA puts the point harder than we would dare to, with its own table of ten human foods commonly fed to cats, and the note that for many cats, consuming just one of these food types in the specified quantity could exceed their maximum recommended daily calorie intake from treats. One ounce, 28 g, of lean cooked skinless chicken breast is 47 kcal, which is more than two days of a 4 kg cat's allowance. The same weight of cheddar is 120 kcal, which is five and a half days. Two fluid ounces of semi-skimmed cow's milk is 27 kcal, which is already over the line.

Why the cat number is so much smaller than the dog number

The gap is the size of the animal rather than anything about feline metabolism. Our guide to homemade dog treats in India works the same ceiling for a dog and lands on 28 to 35 kcal at 5 kg and 47 to 59 kcal at 10 kg. A 4 kg cat gets 22. Per kilogram of bodyweight the two species are not far apart; the difference is that the median Indian house dog is two to five times the weight of the median Indian house cat, so the cat's absolute budget is a fraction of the dog's while the pinch of food a human hand picks up is the same size in both houses.

There is a second, smaller difference and it matters at the margins. A dog treat allowance of 50 kcal tolerates a bad guess by five kilocalories. A cat allowance of 22 does not. The practical consequence is that the cat page has to be about piece size and arithmetic from the first paragraph, where the dog page can get away with starting at ingredients.

Everything in this allowance comes out of the day's food rather than going on top of it. If the treats add up to 22 kcal, the bowl loses 22 kcal. Our cat diet chart for India gives the daily grams to subtract from, and our condition page on obesity in dogs and cats covers how a vet tightens the budget further for a cat that is already heavy.

What a treat can and cannot do for an obligate carnivore

A treat cannot close a nutrient gap in a cat's diet and it can overshoot a nutrient ceiling, and the reason is concentration. No kitchen ingredient is more than about twice as dense as a cat's taurine requirement per calorie, while goat liver is 498 times as dense as its vitamin A minimum. One of those arithmetic facts makes a treat useless as nutrition and the other makes it capable of real harm.

Start with the direction everybody assumes. A reader who has heard that cats need taurine reasonably wonders whether a homemade treat should be built to supply some, and the answer is that it cannot, however good the ingredient. FEDIAF's minimum for a wet or heat-processed adult cat food is 0.50 g of taurine per 1,000 kcal of metabolisable energy, which is 112.5 mg a day for a 4 kg cat eating 225 kcal. The densest taurine source in the UC Davis survey of animal feed ingredients is marine, and among the meats an Indian kitchen actually buys it is chicken dark meat at 1,690 plus or minus 370 mg per kg wet weight, which is 169 mg in every 100 g. Sixty-seven grams of chicken thigh covers a day's taurine. The whole treat allowance buys between 13 and 24 grams of meat or offal. A treat is roughly three times too small to matter, and that is using the best ingredient available.

Now the other direction. India's own food composition tables measure goat liver at 15,655 plus or minus 972 micrograms of retinol per 100 g across six regions, and they state their own conversion, that one international unit of vitamin A equals 0.3 micrograms of retinol. That makes goat liver 415 IU in every kilocalorie. FEDIAF's adult cat minimum is 833 IU per 1,000 kcal, which is 0.833 IU per kilocalorie. Goat liver is 498 times as concentrated in vitamin A as the cat's own requirement, measured the way a treat is actually spent, which is per calorie.

Boiling the chicken throws the taurine down the sink

Boil a cat's meat and most of the taurine leaves with the water. Spitze and colleagues measured the same ingredient raw and cooked, and a chicken leg went from 337 mg of taurine per kg raw, to 229 after baking, to 82 after boiling, so boiling retained 24 percent of it. Their explanation is a property of the molecule rather than of chicken. Taurine has high water solubility, most of the taurine in tissue dissolves into water if exposed, and loss increases through the order raw, fried with juices retained, fried without juices, baking, then boiling, with boiling the worst because the food is surrounded by water. Unless the water is included in the meal, the animal is consuming less taurine than predicted.

Two practical rules fall out of that, and they are free. Bake or dry rather than boil, if the thing you are making is meat. And where you do boil, keep the water and pour it over the food instead of draining it, which is the one case where an Indian kitchen's habit of saving the stock is exactly right.

Their other finding closes the vegetarian question before it opens. All plant products tested in that laboratory contained undetectable levels of taurine. There is no flour, pulse, vegetable or fruit treat that supplies a cat anything it is short of, which is a separate matter from whether it is safe. Our guide to Indian food safe for cats sets out what the Indian plate delivers against the published feline minimums, and our guide to homemade cat food in India sets out what a complete bowl has to carry. Neither job belongs to a treat.

The dark meat figure is worth carrying into the kitchen on its own. Chicken dark meat carries 1,690 mg of taurine per kg against light meat's 180, which is nine times as much. So the thigh and the leg beat the breast for a cat, which is the reverse of what a human cook assumes, and the cheaper cut is the better one.

Liver, and the number that changes for a 4 kg cat

Liver is the one ingredient on this page with a hard ceiling attached, and the ceiling for a cat sits six times lower than for a dog, because 0.40 g of goat liver covers an adult cat's whole daily recommended allowance of vitamin A against 2.4 g for a dog.

The composition figures are measured, not estimated. The ICMR-National Institute of Nutrition's Indian Food Composition Tables, compiled in Hyderabad, put goat liver at 15,655 plus or minus 972 micrograms of retinol per 100 g across six regions, chicken liver at 3,486 micrograms from a single region, chicken gizzard at 46.78 micrograms and goat tripe at 2.17 plus or minus 0.52 micrograms. Goat liver carries four and a half times the retinol of chicken liver and three hundred and thirty-five times that of gizzard. In India that distinction has a shop attached to it, because the kaleji at a butcher's counter is usually goat while the cheap tray at a supermarket is usually chicken.

Set those against the requirement. The National Research Council's daily recommended allowance of vitamin A for an adult cat is 63 micrograms, footnoted as the daily needs for an adult cat weighing 9 pounds consuming 250 Calories per day. Four tenths of a gram of goat liver supplies all of it. One point eight grams of chicken liver does the same. Our guide to homemade dog treats in India runs the identical sum off the companion dog booklet and lands on 2.4 g of goat liver for an adult dog, so the cat's figure is six times smaller, and that is the single clearest reason the cat answer is not the dog answer.

Read the gizzard row twice, because it is the most useful line on this page and it appears nowhere else. Chicken gizzard buys you 23.8 grams of treat for the same 22 kilocalories that buy 17.5 grams of goat liver, so it is the most generous treat on the list by volume, and it carries one three-hundred-and-thirty-fifth of the retinol. A cheap Indian offal most kitchens already throw away is the one that lets you be slightly generous without approaching any ceiling.

Where the cat authorities agree, and why the agreement is the surprise

The two bodies that publish cat ceilings land within 20 percent of each other, which was not what we expected. AAFCO's cat food nutrient profiles set a vitamin A maximum of 333,300 IU per kg of dry matter, which is 83,325 IU per 1,000 kcal of metabolisable energy. FEDIAF sets a nutritional maximum of 40,000 IU per 100 g of dry matter for adult cats and growth, which is 100,000 IU per 1,000 kcal. For dogs the equivalent figures are eightfold apart, as our dog treats page sets out, so the cat question is the settled one.

The surprise is the direction. FEDIAF's vitamin A nutritional maximum for a cat, 40,000 IU per 100 g of dry matter and 100,000 IU per 1,000 kcal, is exactly the same figure it sets for a dog, while its adult cat minimum of 833 to 1,110 IU per 1,000 kcal is roughly half the adult dog minimum of 1,515 to 1,754. AAFCO's cat maximum of 83,325 IU per 1,000 kcal is higher than its dog maximum of 62,500. So the species that actually gets the disease needs less of the vitamin and is given an equal or wider permitted window than the species in which the toxicity is rare. That is not an error, it is what the kitten feeding data support, but it means you cannot read safety off the ceiling by comparing the two animals.

The study the cat ceiling rests on, and the gap it does not cover

FEDIAF names its evidence, and it is a single kitten study from 1967. Its maximum of 40,000 IU per 100 g of dry matter is about 50 percent of the maximum no-observed-adverse-effect level reported by Seawright and colleagues in kittens from 6 to 8 weeks of age fed for 41 weeks, and that study found no adverse effects at a vitamin A intake of 50,000 IU per kg of bodyweight, corresponding to about 90,000 IU per 100 g of dry matter. FEDIAF's reasoning for extending a kitten figure to adults is stated plainly, that since kittens are at least equally vulnerable as adults to hypervitaminosis A, this level should also be safe for adult cats.

One life stage breaks that pattern completely, and the gap in the data is fiftyfold. Freytag and colleagues reported in 2003 that feeding a food with 100,000 IU per 100 g of dry matter to pregnant queens caused fatal malformations in kittens, and that the next lowest value tested, 2,000 IU per 100 g of dry matter, caused no adverse effects. Nothing between those two levels has been tested. From those data the NRC recommended not exceeding 33,330 IU per 100 g of dry matter in food intended for reproduction, and FEDIAF adopted the same figure for products designed for reproducing queens.

The conservative rule for an ordinary adult cat follows from the lowest published numbers rather than from the highest. Treat goat liver as a few grams a week rather than a few grams a day, make chicken liver or gizzard the everyday option, and remember that vitamin A accumulates, so the question is what you do every week and not what happened once. Our page on supplements for dogs and cats in India covers the other route into the same ceiling, which is a cod liver oil or multivitamin given on top of a complete food that was already formulated to sit under it.

What people are actually asking is visible in the search box. Google India's related searches for "can cats eat liver" on 31 October 2026 were "Can cats eat chicken liver everyday", "Can cats eat beef liver everyday", "How often can cats eat chicken liver", "How to cook chicken liver for cats" and "Is beef liver good for cats";. Every one of those is a frequency question, and not one of the pages answering them gives a gram figure against a cat's energy intake.

What hypervitaminosis A does to a cat, and how long it takes

Hypervitaminosis A in a cat builds new bone around joints and vertebrae over months to years, starts with lameness and stiffness before the bone is visible, and does not meaningfully reverse once established. The condition's existence is already covered on our homemade cat food and supplements pages; what follows is the frequency and the clock, which neither answers.

The chronic form is the feline one. It was first reported as a debilitating metabolic osteopathy, deforming cervical spondylosis, associated with diets based solely on milk and raw ruminant, porcine or chicken liver, and the cat is particularly susceptible to vitamin A toxicity with sequelae that are quite unique and distinctive from those seen in other species. Milk and liver is not a historical curiosity in India. It is what a great many households still put in front of a cat, and our page on whether cats can drink milk deals with the other half of that bowl.

The timing is the part owners need. Most natural cases involve adult cats aged 2 to 9 years with no breed or sex predilection, the effects are established after months or years of excess intake, and the characteristic radiographic findings may be detected after 10 weeks on an induction diet. Ten weeks is the shortest published interval between starting and seeing it on a film. That is a season of treats, not a lifetime.

One case shows what that looks like in a flat. A 9-year-old male castrated domestic shorthair that lived strictly indoors and was fed exclusively a home-made diet based on raw pork liver developed lameness of one forelimb that progressed to paralysis over two months, with flaccid paralysis and muscle atrophy of the whole limb and an ipsilateral Horner's syndrome. Radiographs showed massive new bone formation on the ventral aspect of the second cervical to the sixth thoracic vertebra, displacing the trachea. Its serum vitamin A was 630 micrograms per decilitre against a reported feline normal range of 50 to 200, and cats receiving liver diets commonly reach 451 to 1,281.

That cat recovered, which is the unusual part and the reason the case was published. It regained motor function over about six months after the diet was changed to a balanced commercial canned food. The general prognosis is worse. In chronically and severely affected cats, correcting the diet stops the lesions progressing but produces no significant reversal or improvement of the clinical signs.

• Early signs come before the bone. Lameness of one or both forelimbs, stiffness and reluctance to move indicate pain and precede the establishment of bone lesions.
• A cat that has stopped grooming its own back. Cervical immobilisation prevents self-grooming and, in severe cases, prehension of food, so an unkempt coat over the shoulders is a mechanical sign and not a laziness one.
• An odd sitting posture. Affected cats often adopt a marsupial-like posture to compensate for being unable to move the head and neck.
• A change of voice, or regurgitation. Less common findings, caused by exostoses compressing laryngeal structures or the oesophagus.
• Blood work that looks normal. Apart from the raised serum vitamin A, most clinical cases show no other haematological abnormality, including alkaline phosphatase, calcium and phosphorus.

The kitten form, and the part that does not come back

A different form occurs in kittens and some of it is permanent. It includes oedema and inflammation of the gingivae, loosening or retention of the incisor teeth, and damage to the epiphyseal growth plates of the long bones, where vitamin A inhibits chondrocyte multiplication; the resulting reduction in longitudinal growth is irreversible even with cessation of excess vitamin A intake, so the shaft thickness recovers but the proportions do not. The NRC's own framing of the risk is a kitten one, that too much vitamin A is a natural consequence of feeding large amounts of liver to growing kittens and causes skeletal lesions, particularly outgrowths of the cervical vertebrae.

So a liver treat is a worse idea for a kitten than for an adult, which is the opposite of how treats are usually rationed in an Indian home, where the kitten gets the indulgences. Our kitten care guide for India covers the growth diet; the treat rule for a kitten is the same 10 percent of its own much higher energy requirement, spent on muscle meat.

The calcium red herring

Do not try to fix a liver habit with calcium. Contrary to the suggestion that the development of exostoses in hypervitaminosis A may depend on low calcium and high phosphorus ratios, the amounts and relative proportions of those elements in the diet had little or no influence on the development of lesions. Calcium and phosphorus balance is a real and separate problem for a home-cooked bowl, and our guide to balancing a home-cooked diet covers it. It is not a defence against vitamin A.

What makes a treat work on a cat, which is neither sweetness nor masala

A treat works on a cat through umami, and the receptor that detects it responds to purine nucleotides from animal tissue, not to the glutamate a human cook reaches for. Cats cannot taste sweetness at all, which our guide to Indian food safe for cats establishes from the pseudogene; this section is the forward half, which is what a cat can taste and what that means for a treat.

The receptor work is recent and it is unambiguous. McGrane and colleagues propose that umami, mediated by the Tas1r1-Tas1r3 heterodimer, is the main appetitive taste modality for the domestic cat, and showed for the first time that cats express Tas1r1 in taste papillae. The cat umami receptor responds to a range of nucleotides as agonists, with the purine nucleotides having the highest activity, and it does not respond to any amino acids alone; instead 11 l-amino acids act as enhancers in combination with a nucleotide.

Then comes the line that should change what you put in a cat treat. L-glutamic acid and l-aspartic acid are not active as either agonists or enhancers of the cat umami receptor, because of changes in key binding residues at positions 170 and 302. Glutamic acid and its sodium salt, monosodium glutamate, are the compounds the human umami receptor responds to. The savoury compound a human cook uses to make something taste more of itself is the one umami compound a cat's receptor cannot read.

The work also explains the one food every Indian cat owner already knows about. The authors postulate that the renowned palatability of tuna for cats is due, at least in part, to its specific combination of high levels of inosine monophosphate and free l-histidine, which produces a strong synergistic umami taste enhancement. The reported IMP content of yellowfin and albacore tuna is around 8.4 to 10 mM and the free l-histidine content of yellowfin and skipjack 7.9 to 89.5 mM, and both are the highest-concentration compound of their class in tuna, with white meat higher in free histidine than red. L-histidine was significantly preferred to every other amino acid in all comparisons in their behavioural test.

One further detail makes the whole thing tidy rather than merely interesting. Five of the eleven essential amino acids for the domestic cat, l-histidine, l-leucine, l-methionine, l-phenylalanine and l-tryptophan, are also enhancers of the cat umami receptor. What tastes strongest to a cat and what the cat is built to need overlap by design, and the authors suggest the receptor may act as a signal for protein quality.

The practical rule this gives you

Make the treat out of plain animal tissue and add nothing. That is the whole instruction, and it now has a receptor behind it rather than a convention. Three corollaries are worth stating.

1. Serve it warm rather than cold. Temperature is a separate palatability lever with its own evidence, and our guide to fussy eating and appetite loss carries the measured result for cats over seven.
2. Do not reach for cheese, curd or milk to make it tastier. WSAVA's cat list names milk and dairy products among the ingredients to avoid, noting that many cats are lactose intolerant, and the calorie cost of a spoon is worked out on our page about milk for cats.
3. Treat fish as a rotation and not a routine. Tuna is the most umami-dense thing you can hand a cat and it brings its own problems. Our guide to fish for dogs and cats in India holds the thiaminase evidence, the mercury limits and the salt figure for dried fish, including the Indian legal floor of 12 percent sodium chloride in dry salted fish, so we do not restate any of it here.

WSAVA adds one cat-specific instruction that has nothing to do with calories, which is to ensure that any treat offered is an appropriate size and shape to reduce the risk of choking, and to supervise the cat with chews and treats. A cat swallows a small object more readily than it chews one. The oesophageal foreign body literature for cats is covered on our fish page, and the practical reading for a homemade treat is to make it crumble rather than to make it a disc the cat can take whole.

Four methods, given as techniques rather than as recipes

Use four techniques rather than four recipes, because a recipe fixes a quantity in cups and a cat's whole treat allowance is measured in single grams. Our dog page refuses recipes for the same reason and the argument is stronger here, because the two ingredients a cat recipe reaches for, liver and tinned fish, are the two whose concentration can overshoot a ceiling.

The recipes ranking in India are the proof. The highest-ranking homemade cat treat recipe that gives a liver quantity calls for half a pound of raw beef liver in one batch, alongside two eggs, four tablespoons of vegetable oil, three tablespoons of pumpkin and a cup of instant oats, and its only quantity guidance for the cat is "remember to treat responsibly; treats have calories too" with no figure attached. India's own tables do not analyse beef liver, so no Indian figure exists for it, and we are not going to invent one. What can be said is that half a pound is 227 grams, and the same weight of goat liver would carry about 35,500 micrograms of retinol, which is roughly 564 days of a 4 kg cat's recommended daily allowance of 63 micrograms National Research Council. That is a recipe, in a cup measure, holding about eighteen months of one nutrient.

1. Dry a single ingredient. Slice chicken gizzard, chicken thigh or chicken liver thin, lay the pieces on a rack, and hold them in the oven at its lowest setting until they are brittle rather than leathery. One ingredient means you know exactly what is in it, drying keeps the taurine where boiling would leach it out, and a brittle sheet snaps into crumbs that are naturally the right piece size. Gizzard is the one to default to, because it is the lowest-retinol option in the table above.
2. Break a crumb at the moment. For a short session, tear up a piece of plain cooked chicken thigh as you go. Nothing is stored, nothing can go mouldy, and the only arithmetic is the weight of what you started with. This is the method for a humid month.
3. Freeze a lick. Purée plain cooked chicken with a little of its own cooking water and freeze it in an ice tray, then offer a part-thawed cube. The lickable format is what the Indian market actually sells, and it suits a cat that takes food with its tongue rather than its teeth. Keeping the cooking water is the point, since that is where the taurine went.
4. Spend the allowance on nothing at all. A plant reward costs zero kilocalories and works on most cats. Bol and colleagues tested catnip, silver vine, Tatarian honeysuckle and valerian root on 100 domestic cats: 94 of the 100 responded to at least one of the four, with silver vine at 79 percent and catnip at 68 percent, and 6 cats responded to none.

That fourth method deserves its own paragraph, because it is the only way out of a 22 kilocalorie budget. Of the 31 cats that did not respond to catnip, 22, which is 71 percent, responded positively to silver vine, 10 to Tatarian honeysuckle and 6 to valerian root. So a cat that ignores catnip is not an unrewardable cat, it is a cat that has been offered one of four options. The same study asked the professionals and found that 36 of 38 veterinarians, 95 percent, and all six veterinary technicians were unaware of silver vine or Tatarian honeysuckle and their effects on cats, so do not expect to be told about it at a clinic. For what to do with the response once you have it, our guide to things to do with your cat in India covers play and the clicker method, and indoor cat boredom and enrichment covers food puzzles and the DIY feeders an Indian kitchen can make from what it throws out.

How small is small, for a cat

Half a gram is the training piece, and at that size the allowance stretches. A 0.5 g crumb of dried chicken gizzard is 0.46 kcal, so 47 of them fit inside a 4 kg cat's 22 kcal day; the same crumb of chicken liver is 0.62 kcal, so 35 fit; of chicken breast, 0.84 kcal, so 26 fit. A single 3 g cube of any of them is a seventh to a tenth of the whole day. The way to hit half a gram is to make the treat in a form that breaks small rather than to cut thirty pieces with a knife, which is why a dried brittle sheet beats a moulded ball.

That number is the one the ranking pages replace with advice. The best of the product listicles says the right training treat is a soft single-ingredient reward of around 3 to 4 calories each. Six of those is a 4 kg cat's entire day. A training session is thirty to forty repetitions. The arithmetic does not survive contact with its own recommendation.

Storage follows from the batch size rather than from a shelf-life rule. A week of treats for a cat is a handful, which fits in the refrigerator, and in Mumbai, Chennai and coastal Karnataka and Tamil Nadu the humid season runs for months rather than weeks, so the freezer is the default for anything beyond a few days. Our guide to storing pet food in Indian humidity covers the mechanism, and our dog treats page carries the aflatoxin evidence for a grain-based treat, which a cat treat avoids by not containing grain in the first place.

The treat risk the research actually found, which is not the treat

The one prospective cohort to test treat practice against feline weight found that the occasion mattered and the frequency did not. Owners who gave their cat treats when they "felt happy" with them carried an odds ratio of 2.7 for overweight at two years, while treat frequency, tested in the same cohort at one year, did not survive into the final model at all.

Both results come from the Bristol longitudinal studies of pet cats. In the one-year analysis, cat breed, neuter status, outdoor access, type of diet, frequency of wet and dry food fed and frequency of treats fed were all analysed as potential risk factors in 966 cats, of which 7.0 percent were reported overweight or obese at 12.5 to 13 months. The final multivariable model identified two risk factors, restricted or no outdoor access and feeding dry food as the only or major part of the diet, and treat frequency was not among them.

The two-year analysis is where treats reappear, in a different shape. Of 375 cats with owner-reported body condition scores, 25.3 percent were overweight or obese at two years, and six factors were significantly associated with it. One of those six was owners giving their cat treats when they felt happy with them at 18 months of age, at an odds ratio of 2.7 with a 95 percent confidence interval of 1.0 to 7.3. That interval touches 1, so the association is marginal, and we are printing the interval rather than the point estimate alone for exactly that reason.

Why how often turned out to be the wrong question

A frequency rule has no ceiling, and that is the mechanism these two results point at. A cat given one crumb a day and a cat given a quarter of its day's energy in treats daily both answer "daily" on a questionnaire, which is why frequency was a weak variable. The variable that survived describes an occasion that has no natural limit, because there is no upper bound on how often an owner feels pleased with a cat. A treat given to mark a behaviour is countable and a treat given to express affection is not.

This matters in India because the frequency rule is the only rule the Indian pages give. The one Indian article in the top seven results for homemade cat treats says that rewarding them with treats once or twice a week is considered ideal, with no quantity, no weight and no calorie figure anywhere on the page. That is the variable the Bristol cohort tested and failed to retain, published as the answer.

Two honest limits on all of it. Both studies used owner-reported body condition scores in a United Kingdom cohort, and no Indian study has measured treat practice or feline body condition at all, so the odds ratios are a signal about how treating behaves and not a figure about Indian cats. And neither study measured treat calories, which remains unmeasured anywhere.

What a cat treat costs in India, homemade against bought

A 4 kg cat's whole daily treat allowance costs ₹3.24 as chicken liver, which is about ₹97 a month, against ₹3.71 for a single gram of an imported-style single-ingredient treat. The as-sold gap is twentyfold and the honest gap is closer to sixfold, because one of the two is mostly water.

The homemade side is one price and one division. A 450 g tray of chicken liver was Rs 82 on BigBasket on 31 October 2026, which the seller itself states as ₹18.22 per 100 g. A 4 kg cat's 22 kcal allowance is 17.8 g of chicken liver, so a full day of treats costs ₹3.24 and a month costs about ₹97. A week's worth is 125 g, about a quarter of the tray, and the rest goes in the freezer.

That figure is a Bengaluru-headquartered national retailer's price, and a neighbourhood butcher in Delhi, Pune, Chennai or Kolkata usually sells liver and gizzard below it, so re-run the sum with a local price rather than quoting ours. The shape of the answer does not change. Offal is the cheapest animal protein a cat can be given and the arithmetic stays trivial.

The dry matter column is the one to read, and it is why the obvious comparison is misleading. Fresh chicken liver is 73.20 percent water by India's own measurement, and the dried pack states a maximum moisture of 8 percent, so comparing them by the hundred grams as sold compares a hundred grams of liver with twenty-seven grams of dry matter against a hundred grams of fish with at least ninety-two. On the as-sold figures the commercial treat is 20 times the price. On dry matter it is about six times. Six times is still six times, and homemade wins the cost question comfortably, but we are not going to publish the flattering number when the honest one is available.

Two things on that pack are worth naming as gaps rather than as faults. Its feeding instruction for a cat is two to three sprats a day with no calorie figure, no gram weight and no percentage, so a reader cannot place it inside the 10 percent ceiling; WSAVA's own instruction for exactly that situation is to contact the manufacturer for calorie information if it is not stated on the label Dutch Hollandia (Indian retailer) product…. And it carries the claim "Vet-approved" without naming a veterinarian.

One cost comparison does not work and it is worth saying so. A homemade treat has no manufacturing, packaging or shelf-life cost, and it also has no calorie declaration, no shelf-life test and no quality control, so the price difference is not purely a mark-up. What you are buying with the extra rupees is a pack that keeps in a humid cupboard and a weight on a label. What you are giving up is the one thing a 22 kilocalorie budget needs most, which is knowing exactly what went in.

A treat is not a meal, and nothing on this page is a diet

Treat nothing in this guide as nutrition. WSAVA is explicit that treats should never replace a meal and that every cat needs a complete and balanced diet, and for an obligate carnivore that sentence has more weight than it does for a dog.

The two questions get confused constantly, and for a cat the consequences of confusing them are larger. A homemade meal for a cat has to meet a published nutrient profile across dozens of nutrients, four of which the animal cannot make at all, and our guide to homemade cat food in India sets out those minimums and what a home recipe actually delivers against them. A treat has the opposite job. It is deliberately incomplete, it exists to pay for a behaviour or to carry a tablet, and it is capped at 10 percent precisely so that its incompleteness cannot matter.

Ten percent of the day cannot unbalance a diet. Thirty percent can, and for a cat eating 225 kcal, thirty percent is 67 kilocalories, which is three crumbly cubes of dried liver nobody counted. That is the whole risk, and it is an arithmetic risk rather than an ingredient one.

Some cats need the allowance cut or dropped, and that is a veterinary decision rather than a reading of this page. A cat being managed for weight carries a tighter budget than 10 percent, and our page on obesity in dogs and cats covers how it is set. A cat with chronic kidney disease has a phosphorus and protein ceiling that rules out offal treats. A cat with feline lower urinary tract disease or diabetes may be on a prescription diet that conventional treats undo, which WSAVA names directly, noting that where a cat has a medical condition, dietary sensitivities, or is fed a veterinary-prescribed diet, conventional treats may not be appropriate. Ask your own veterinary surgeon before adding anything to such a cat's day.

Community cats sit outside all of this and the honest answer for them is different. A free-ranging cat has no ration to subtract a treat from, so the 10 percent rule has nothing to attach to, and the useful thing to put down is food rather than treats. Our guide to community cats in India covers what to feed and the civic position.

Finally, the list of what never goes into a cat treat is WSAVA's rather than ours. Its toxic ingredients for cats are onions, leeks, shallots, garlic, chives, grapes, raisins, currants and sultanas, chocolate, xylitol, alcohol and raw yeast dough, caffeinated drinks including coffee, tea and energy drinks, and milk and dairy products, the last because many cats are lactose intolerant. The Indian form of the first item is the ginger-garlic base under every gravy, and the dose figures for both species are on our onion and garlic page. We do not restate toxicology here, because this is a page about arithmetic.

What this page can prove, and what it cannot

Separate the measured from the inferred, because a page telling an Indian owner to feed a cat something out of their own kitchen is exactly where unsourced confidence does damage.

Proven, with the source beside it.

• Treats must stay under 10 percent of a cat's daily calorie intake and must never replace a meal, which is 18 kcal a day at 2.5 kg and 22 kcal at 4 kg, against a daily requirement of 225 to 250 kcal for a 4 kg cat.
• Goat liver carries 15,655 plus or minus 972 micrograms of retinol per 100 g across six Indian regions, chicken liver 3,486 from one region and chicken gizzard 46.78, and 0.40 g of goat liver supplies an adult cat's whole daily recommended allowance of 63 micrograms.
• AAFCO and FEDIAF set the cat's vitamin A ceiling at 83,325 and 100,000 IU per 1,000 kcal, and FEDIAF's cat figure is identical to its dog figure while the cat's minimum is about half the dog's.
• Kittens fed for 41 weeks at about 90,000 IU per 100 g of dry matter showed no adverse effect, and 100,000 IU per 100 g of dry matter fed to pregnant queens caused fatal malformations in their kittens.
• Boiling a chicken leg left 82 mg of taurine per kg against 337 raw, and no plant product tested contained detectable taurine.
• The cat umami receptor responds to purine nucleotides and not to glutamic or aspartic acid either as agonists or as enhancers.
• Ninety-four of 100 cats responded to at least one of catnip, silver vine, Tatarian honeysuckle and valerian root.

Not proven, and recorded here as unknown rather than quietly left out.

• No feeding trial has given a cat goat liver at a measured dose, so every liver figure on this page is Indian composition data applied to a published whole-diet ceiling rather than a feeding result.
• No published case of feline hypervitaminosis A states the dietary intake in international units, so there is no gram-per-day threshold for a cat, and anybody who gives you one has made it up.
• Nothing between 2,000 and 100,000 IU per 100 g of dry matter has been tested in a pregnant queen, a fiftyfold untested gap, which is why the reproduction ceiling is a conservative extrapolation rather than a measured threshold.
• The chicken liver and gizzard retinol figures come from a single region each, while the goat liver figure is a six-region mean with a stated error, so the three do not carry equal weight.
• India's tables do not analyse beef liver, so no Indian retinol figure exists for the ingredient the ranking recipes actually use.
• The Bristol treat association has a confidence interval of 1.0 to 7.3, which touches 1, and no Indian cohort has measured treat practice or feline body condition at all.
• The umami work measured receptor activity and short-term intake, not training performance, so it does not tell you which treat a cat will work hardest for; no equivalent of the canine reward-size experiment has been run in cats.
• Nobody has measured the calorie content of a homemade cat treat, in India or anywhere, which is why this page keeps telling you to weigh what goes in rather than to weigh what comes out.

Where the authorities agreed we said so and explained why the agreement was the surprise. Where the evidence cut against this page's own argument, as the one-year Bristol model does by failing to retain treat frequency, we published it in the section it complicates rather than at the bottom.