Dog Food Calculator

How much to feed your dog, by calories and cups — based on weight and life stage. Updated for 2026.

Last reviewed September 2026 · Free · Nothing you enter is stored

How much should I feed my dog?

The honest answer is that it depends on weight, age, neuter status, activity, and the specific food's calorie density — and that even with all of those, the result is a starting point rather than an answer. This calculator does the veterinary arithmetic: it finds your dog's Resting Energy Requirement, multiplies by a life-stage factor to get the Maintenance Energy Requirement, and divides by your food's calories per cup. That is a far better place to begin than the feeding chart on the bag, which knows nothing about your particular dog.

What follows is where each of those numbers comes from, and how much confidence it deserves. Most feeding calculators present a single figure with no error bars. The error bars here are wide enough to matter.

The resting requirement, and why the shortcut version is worse

Resting Energy Requirement is the energy a dog burns lying still in a comfortable room, digesting nothing in particular. Every veterinary source uses the same allometric equation for it: RER = 70 × (body weight in kg)^0.75. The three-quarter exponent reflects the long-observed fact that metabolic rate does not scale with body mass in a straight line — a 40 kg dog does not burn twice what a 20 kg dog burns.

A linear shortcut, RER = (30 × kg) + 70, circulates widely because it can be done on a clinic calculator without an exponent key. It should not be used on a website, where the real equation is free. AAHA describes the shortcut as valid for dogs between 2 and 25 kg and says it "will overestimate" outside that band; Merck gives the range as 2 to 45 kg. Both are describing the same approximation, and neither description is quite right — worked against the exact equation, the linear form actually underestimates by around 6% for dogs between roughly 4 and 18 kg, and overestimates by 17% at 45 kg. This calculator uses the allometric equation throughout.

The life-stage multipliers, and the disagreement nobody mentions

The multiplier table used by essentially every online calculator — 1.6 for a neutered adult, 1.8 intact, 3.0 for a puppy under four months, 2.0 from four months to adult — traces to a single veterinary textbook, Small Animal Clinical Nutrition. Merck publishes a shorter list that agrees with it exactly where the two overlap. So far, so consistent.

The complication is that the World Small Animal Veterinary Association does not use that method at all. Its owner-facing calorie charts apply the NRC 2006 direct equations instead: 130 × kg^0.75 for an active adult dog, 95 × kg^0.75 for an inactive one. For a 10 kg neutered dog, the two approaches give materially different answers — 630 kcal by the multiplier method, 534 by the NRC inactive equation, 731 by the NRC active equation, and 470 to 590 according to the table WSAVA prints in the same document. That is a spread of 470 to 731 kcal a day for one ordinary dog, from sources that are all authoritative.

We use the multiplier method here because it is the one that accommodates life stage, activity and weight-loss goals in a single consistent framework. We are not presenting it as more correct than the alternatives. Anyone who tells you there is one right number for a dog of a given weight has not read the primary sources.

How wrong the estimate can be

Both major authorities publish an explicit uncertainty band, and it is much wider than the differences between the multipliers. AAHA's nutritional assessment guidelines state that energy requirements "can vary by 50% in either direction for cats, and by 30% in either direction for dogs." Merck says an individual animal "may require as much as 30% more or less of the calories calculated for its body weight than another animal with the same body weight."

So a calculated 800 kcal per day honestly means somewhere between about 560 and 1,040. That is not a failure of the method; it is the method working as intended. The output is a hypothesis to test. Feed it for two to four weeks, re-weigh, re-score body condition, and adjust by 10% at a time. The dog is the instrument, not the equation.

A cup is a measure of volume, not of food

The largest avoidable error in home feeding is not the equation — it is the cup. A peer-reviewed survey of 93 commercial weight-management diets found dry dog foods ranging from 217 to 440 kcal per cup, with a median of 301. Kibble density varies with piece size, extrusion and fat coating, so two foods with identical calories per kilogram can differ substantially per cup. Fat carries 8.5 kcal per gram against 3.5 for protein and carbohydrate, which moves energy density further still.

Then there is the scooping itself. Research on how accurately owners measure kibble with a household cup found errors ranging from an 18% underestimate to an 80% overestimate; a later study found a spread from 48% under to 152% over, with the worst accuracy at small volumes. Combine a 2× range in calorie density with a 2.5× range in scooping accuracy and the "one cup twice a day" instruction stops meaning anything at all.

The fix costs about fifteen dollars: weigh the food in grams on a kitchen scale. Every compliant bag carries a calorie content statement in kcal per kilogram, which converts to grams directly with no density guesswork. This single change removes more error than any refinement to the formula.

Feeding for weight loss starts from the weight you want

If your dog needs to lose weight, the calculation must run from target weight, not current weight — and this is where most feeding advice quietly fails. The same survey of weight-management diets found that label feeding directions specified, at the median, exactly 1.00 × RER for the dog's current body weight. Feeding that will maintain an overweight dog at its overweight body weight indefinitely, which is presumably not what the buyer intended.

AAHA's guideline is to feed 80% of RER calculated at ideal weight, targeting a loss of 1–2% of body weight per week for dogs. It also warns that more aggressive restriction, below roughly 60% of RER, raises the risk of nutritional deficiency. This is the setting where veterinary involvement genuinely matters, for three reasons: the ideal-weight figure is itself an estimate, restriction below a certain point causes harm, and unexplained weight gain can be a clinical sign — hypothyroidism and hyperadrenocorticism both present this way — rather than a feeding problem.

Body condition score does the work that weight cannot

A number on a scale cannot tell you whether a dog is carrying too much fat, because breeds and frames differ. Body condition scoring on the 9-point scale can: 4–5 is ideal, 1–3 under, 6–9 over. AAHA's rule of thumb is that each point above 5 corresponds to roughly 10% overweight, so a dog at 7 is about 20% over its ideal weight.

Worth knowing that AAHA's own published example is internally inconsistent here. It describes a 45 kg Labrador at body condition 8 as "30% overweight" with an ideal weight of "approximately 32 kg" — but 45 ÷ 1.30 is 34.6 kg, while 45 × 0.70 is 31.5. Whether "30% overweight" means a markup on ideal weight or a fraction of current weight changes the answer by nearly 3 kg, and the guideline does not say which it means. We use the markup reading, ideal = current ÷ (1 + excess), because it is the one that makes the phrase "30% overweight" arithmetically true.

One caution about self-assessment: in a study of 174 dog owners, only 55% correctly identified their dog's body condition from a written description and 51% from pictures, with 38% underestimating. If the score feels like guesswork, ask your vet to score the dog once and show you what it feels like. It takes a minute and recalibrates everything downstream.

Switching foods, and the schedule nobody can source

The standard advice to transition gradually over 7–10 days is sound in principle and unsupported in detail. Tufts publishes a seven-day schedule: 90/10 on day one, 75/25, then 50/50 for two days, 25/75, 10/90, and fully switched by day seven. Cornell publishes a four-day schedule: 75/25, 50/50, 25/75, done. Two veterinary schools, schedules differing by nearly a factor of two, and no trial data behind either.

What both agree on is the useful part: if loose stool appears, hold at the last tolerated ratio for a day or two rather than pressing forward, and expect that some dogs need four to six weeks rather than one. Treats should stay within 10–15% of daily calories throughout, which is a good deal less than most kitchens assume.

Frequently asked questions

Sources: AAHA Weight Management Guidelines and Nutritional Assessment Guidelines; Merck Veterinary Manual, Nutritional Requirements of Small Animals; WSAVA Global Nutrition Committee calorie charts; Linder & Freeman, JAVMA 2010;236(1):74–77; Yam et al., Vet Sci 2017;4:30; Tufts Petfoodology; Cornell Riney Canine Health Center. Figures we could not verify from a primary source are not published here.