How to set portions on an automatic feeder

By Josh · Reviewed and updated: 2026-09-17

An automatic feeder measures in "portions", and a portion is not a unit. It is whatever the manufacturer decided one click of the rotor should drop. We checked the published specs of widely sold models: one portion is listed as about 5 g on some feeders, about 10 g on others, about 12 g on another, and budget models that measure in "small cups" list 25 to 35 g. Set ten portions and you might be dispensing 50 g or 350 g — a sevenfold difference, from the same instruction. This page is about getting from your pet's daily grams to the right number on your machine.

Why a portion is not a unit

Portion size is a design choice, not a standard. The rotor turns once, a fixed volume falls through, and the manufacturer picks the number of turns that they think matches a typical meal. Across the models we looked at, the number they picked varies by a factor of seven:

Two of those machines would give you a wildly different meal from the same setting, and nothing on either machine tells you so. The word "portion" hides the number that matters.

Read the last two rows again, because they make a different point. Those are two feeders from the same brand, and the brand's own published figures disagree: 10 to 18 g on one, 10 to 15 g on the other. Neither is a fixed number. Both carry the same qualifier — "depending on the density of the feed being used" — which is the manufacturer telling you that the portion changes when you change food, on a machine that has not changed at all.

Everything in that table is what the manufacturers publish. None of it has been independently verified, and none of it describes your machine with your food. The only figure that does is the one you measure in Step 2. That is why Step 2 is not optional.

There is a second layer to this, and the manufacturers say it themselves — in three different ways. One brand's FAQ attaches a caveat to its 12 g figure: the actual weight per portion "may vary slightly depending on the size and density of your specific kibble". Another manufacturer's user manual is blunter: "The difference in shape and size of the dry pet food will cause a variation in the amount of food dispensed per serving." And a third puts the mechanism on the record outright: "Portion size is measured by volume rather than weight. Different brands of kibble may produce slightly different dispensing weights."

None of those three sentences is buried in the small print. All three sit exactly where owners go to find out how much a portion is. That last one is worth reading twice, because it is the entire argument in a manufacturer's own words: the machine measures volume, you feed by weight, and nothing on the machine converts between them for you.

Step 1: get the daily amount in grams

Not in cups, and not in "portions". Grams. A cup is a volume and the feeder's portion is also a volume, which means converting volume to volume loses the density question twice. Grams are the unit that survives.

Our calculator gives the daily amount in grams once you enter your pet's weight, life stage, neutered status, activity level and the calorie content of the food. For a 10 kg neutered adult dog on a typical activity level, eating a food of 350 kcal per cup at 110 g per cup:

Split across two meals, that is about 99 g per meal. Now you have a target the feeder can be measured against, and it is in the same unit as a kitchen scale.

Step 2: weigh what one portion actually dispenses

This is the step that makes the whole thing work, and it takes two minutes. Do it once per food, and again any time you change brands or switch to a different bag of the same brand.

1. Put an empty bowl on a kitchen scale and tare it. Use a scale that reads in grams, not ounces — one gram is about 0.035 oz, and a scale that rounds to the nearest ounce cannot see the differences you are looking for.
2. Set the feeder to dispense a single portion, then run that one portion into the bowl.
3. Weigh it and write the number down. Then run two portions and weigh again. Three portions, weigh again.
4. Divide the differences. If one portion weighs 12 g, two weigh 24 g and three weigh 36 g, your per-portion figure is 12 g and it is consistent. If you get 12 g, 25 g and 34 g, the machine is not dropping a fixed amount and you should use the average and check it more often.

Do this with the food you are actually feeding. The number in the manual was measured with whatever food the manufacturer had on the test bench, and it is not a property of your machine alone.

Step 3: convert grams to portions

For the dog above: 198 g per day, two meals, so 99 g per meal. Divide that by whatever one portion weighs on your machine:

All three machines land within a few grams of the target, which is the reassuring part. The unsettling part is the next step.

The part nobody mentions: your step size

The number above is the nearest whole portion, not one you can always hit. What you can actually set is a whole number of portions, so your real ceiling on accuracy is how coarse one step is. Half a step is the most you can ever be off, because anything further away would be closer to the neighbouring setting:

So a fine-stepping machine is not a luxury feature; it is the difference between being able to hold a target and only being able to aim at it. On the same dog, a 25 g step leaves you around 6% away from where you want to be at best, whereas a 5 g step gets you within about 1.3%.

One condition on that table. It assumes you have done Step 2 and have set the closest whole number of portions — that is the situation in which half a step is the most you can be off. If you have never weighed a portion, you do not know how large your step is, and the error can be considerably larger than half of it. Half a step is a bound you earn by measuring, not one you get by default.

And half a step is only the part you control. The same manufacturer puts the food's own contribution at about ±10% and calls it normal, because kibble varies in size and density from one bag to the next. So the error you are managing has two independent parts — the step you can set, and the food you cannot. The weighing in Step 2 is what keeps both of them small; skip it and the two add together.

And that is the ceiling under ideal conditions, when you have done the weighing and set the machine correctly. The error that actually costs owners weight is a different one, and it is a single click:

The point is not that coarse feeders are dangerous. It is that the smaller your step, the less a slip costs you — and that on any machine, a periodic weigh-in is what catches the slip.

Four ways an automatic feeder ends up overfeeding

• Setting it in cups because the bag gives cups. The bag's cup figure assumes a level measuring cup. A feeder portions by volume too, but its portion is not a cup, so the two volumes never line up. Convert the bag figure to grams once, then work in grams from then on.
• Changing food without re-weighing a portion. A denser kibble puts more grams through the same rotor travel. Nothing on the machine changes, so nothing looks wrong, and the amounts drift. Re-weigh whenever the food changes.
• Treating "portions" as comparable between machines. This is the one that leaks across households: a friend says "mine gets six portions", and you set six on your machine. If theirs dispenses 5 g a portion and yours dispenses 25 g, you have just multiplied their feeding by five.
• A stuck dispenser emptying the hopper. Machines fail in the direction of too much as well as too little. A paddle that has not seated properly against the outlet, or a piece of kibble wedged where the gate closes, can leave the gate open and dispense continuously. This is not a hypothetical: a manufacturer publishes a whole troubleshooting article titled "What should I do if the feeder releases too much food?", and the test it asks owners to run is whether the rotor rotates only once when the feed button is pressed. Jam alerts are standard on the models that have an app. On a feeder like that you do not get a portion error. You get the whole hopper — and an empty hopper is how you find out.

What to check before buying a feeder

If you are choosing one, these five specifications decide whether you can actually hit a target, and all five are published on the product page or in the manual:

One thing worth saying plainly: if you only need a fixed amount at fixed times, a button-controlled feeder is cheaper and more reliable than a connected one. Wi-Fi buys you remote scheduling and notifications. It does not buy you a more accurate portion, and the radio drains the batteries faster than a simple timer does.

One more thing that is not in the table but affects accuracy over time: a dirty feeder under-dispenses. Kibble dust and oils build up in the food path and the chute, and a machine that is never taken apart gradually delivers less than it says it does. If it is awkward to clean, it will be cleaned less often.

When a feeder is the wrong tool

Automatic feeders solve a scheduling problem. They do not solve every feeding problem, and there are four cases where a feeder is the wrong answer:

• Wet food. Almost all of these machines are designed for dry kibble and freeze-dried food only. Moist food sticks to the rotor and spoils in the hopper, and it is usually excluded by the warranty rather than fixed by a setting.
• More than one pet. A feeder dispenses to whoever reaches the bowl. In a two-cat household that means the faster cat can eat both meals, and the slower one loses weight while you believe the schedule is working.
• A prescription or weight-loss diet. When the target is being set by a vet, a 6% step is not a small detail. Measure and adjust in grams, and keep the feeder out of it, or use a model you can set in grams.
• Any cat that has stopped eating. A cat that eats nothing for 48 to 72 hours is at risk of hepatic lipidosis, which is a medical emergency rather than a scheduling problem.
• When the feeder is the only thing feeding your pet. If you are away and the machine jams, runs out of food, loses power or stalls, there is no backup and you will not know until you get home. A feeder is a convenience for scheduled meals when someone is around; it is not a substitute for a sitter or a neighbour looking in. If a feeder failure would mean your pet going 24 hours or more without food, it is the wrong tool for that trip.

If you are not sure whether the amount you have been feeding is right, the body condition score page takes two minutes and tells you more than the bowl does. If you would rather measure by hand, the measuring-without-a-scale page covers the alternatives.

Frequently asked questions

Work out your pet's daily amount in grams →
How to measure without a scale →
How much does a cup of your food weigh? →
Check your pet's body condition score →