What Does “Complete and Balanced” Actually Mean on Dog Food?

Updated September 11, 2026 · first published August 27, 2026Rewritten in full against the current regulations

There is one sentence on a bag of dog food that carries real legal weight. Everything else on the front — premium, holistic, ancestral, performance, human grade — is adjectives. This one is a regulated claim, and if a food does not carry it, that food is not a complete diet.

So it is worth knowing exactly what it promises. And the honest answer is that it promises a floor, not a standard, and the strongest version of the claim rests on six to eight dogs, six months, and four blood tests.

That is not a scandal, and this article is not an argument for abandoning complete and balanced foods. It is an argument for knowing what you are holding.

The short answer

“Complete and balanced” means the food is intended to be the animal’s whole diet, and that the manufacturer has substantiated it by one of exactly two routes.

Route one is a calculation. The formula is compared against a nutrient table. No dog eats the food before the claim is made.

Route two is a feeding trial. A small group of adult dogs eats the food for six months and four blood values are checked at the end.

Most foods use route one. It is cheaper, faster and permitted. And neither route involves a government agency approving anything.

What does complete and balanced mean on dog food?

It means the food is formulated to be fed as an animal’s sole diet, and that the manufacturer has substantiated that by one of two permitted methods: calculating the formula against an AAFCO nutrient profile, or running an AAFCO feeding trial. It is a statement about nutrient content on paper or a six-month tolerance test, not a quality rating, not a measure of digestibility, and not an approval by any government agency. A food without the statement is not intended as a whole diet.

Two routes, and only one of them involves a dog

The FDA states the rule plainly. To carry the claim, a dog or cat food must either

“Meet one of the Dog or Cat Food Nutrient Profiles established by the Association of American Feed Control Officials (AAFCO); or Pass a feeding trial using AAFCO procedures.”

Route one: formulation

The manufacturer calculates the finished formula’s nutrient content against the AAFCO nutrient profile for the intended life stage, using published values for each ingredient. If the numbers clear the minimums and stay under the seven maximums, the food qualifies.

No animal eats the food before the claim is made. Nothing is measured in a laboratory unless the company chooses to measure it. The calculation uses database values for ingredients, and database values are averages, not the drum that arrived at the plant on Tuesday.

This is by far the more common route, and it is not fraud. It is a reasonable engineering approach. But it assumes the ingredients are what the database says, the mixing is even, and nothing is destroyed in processing — and the pillar article on what working dogs actually need goes through what happens between the mixer and the bowl when those assumptions do not hold, and what the dog food recalls actually tell you is what it looks like when they fail outright.

Route two: the feeding trial

This is the stronger claim, and it is the one manufacturers advertise when they have it. A bag saying “Animal feeding tests using AAFCO procedures substantiate that…” has done something a calculated food has not.

It is also much smaller than it sounds.

How is a dog food proven to be complete and balanced?

By one of two methods, and the label tells you which. If it says the food “is formulated to meet” an AAFCO profile, the substantiation is a calculation against a nutrient table and no dog ate the food first. If it says “animal feeding tests using AAFCO procedures substantiate,” the food was fed to a group of dogs for at least 26 weeks. The second is stronger evidence. Neither is an approval, and neither is a quality rating.

The feeding trial, in full

Here is the entire adult maintenance protocol, quoted from AAFCO’s own document:

Read the dropout line again. Eight dogs start; a quarter may be removed. A trial can be passed by six dogs.

Read the weight line again too. A dog can finish this trial fifteen percent lighter than it started and the food passes.

And the laboratory work is four values. Hemoglobin, packed cell volume, alkaline phosphatase, albumin. That is the complete list.

What the trial does not look at

Not because anyone is hiding anything — the protocol is public — but because a 26-week tolerance test in adult dogs cannot see:

• Taurine status. Not measured.
• Vitamin D status. Not measured. Both the deficiency and the toxicity are invisible here.
• Bone density or skeletal development. Not measured, and the trial uses adults, so growth problems could not appear in it anyway.
• Anything that takes longer than six months. Heart disease, liver copper accumulation, kidney effects — all slower than the test.
• Digestibility. Not required.
• Lifespan. Obviously.

To be fair to AAFCO, it says something similar itself. Its own guidance notes that formulating to the profiles is “only one part of a nutritionally sound, scientific development” and that meeting a profile “should not deter or discourage the manufacturer from conducting appropriate feeding trials.” The organization writing the standard is on record that the standard is a starting point.

How many dogs are used in an AAFCO feeding trial?

A minimum of eight adult dogs must start, and up to 25 percent may be removed during the test for non-nutritional reasons or poor food intake — so a trial can legitimately finish with six. It runs at least 26 weeks. Four blood values are measured at the end: hemoglobin, packed cell volume, serum alkaline phosphatase and serum albumin. No individual dog may lose more than 15 percent of its starting body weight. That is the full protocol, and it is the strongest nutritional adequacy claim available on a bag.

Where the sentence lives, and how to read it

The nutritional adequacy statement is in small print, usually near the feeding directions rather than on the front. It has a fixed shape:

“[Name of product] is formulated to meet the nutritional levels established by the AAFCO Dog Food Nutrient Profiles for [life stage].”

or

“Animal feeding tests using AAFCO procedures substantiate that [name of product] provides complete and balanced nutrition for [life stage].”

Three things to take from it.

1. Which route was used. “Formulated to meet” is the calculation. “Animal feeding tests substantiate” is the trial.

2. Which life stage. There are only two profiles — growth and reproduction, and adult maintenance. “All life stages” is not a third option; it means the food meets the growth profile, which is the stricter one. If you have a puppy who will mature over 70 pounds, the statement must also cover growth of large-size dogs, and the calcium ceiling that applies to big puppies is where that gets serious.

3. Whether the statement is there at all. A product labeled “for intermittent or supplemental feeding only” is telling you, in the regulated part of the label, that it is not a diet. Toppers, some raw products, most treats, and a fair number of things marketed like meals carry that line.

Which sentence on a dog food bag matters most?

The nutritional adequacy statement, in the small print near the feeding directions. It tells you whether the food is a complete diet or only a supplement, which life stage it was substantiated for, and which of the two permitted methods was used to substantiate it. Nothing on the front of the bag carries comparable regulatory weight — “premium,” “holistic,” “natural” and “performance” are marketing terms with no nutritional definition behind them.

What does “for intermittent or supplemental feeding only” mean?

It means the product is not a complete diet and must not be fed as one. Any pet food that does not meet an AAFCO nutrient profile or pass a feeding trial has to carry that statement instead of a complete-and-balanced claim, so it is the manufacturer saying in the regulated part of the label that this food is a topper, a treat or a partial ration. Some products that look and are priced like meals carry it. Check before you replace a diet with one.

And nobody is approving any of it

This is the part that surprises people most, and every piece of it comes from the bodies involved.

AAFCO is not a regulator. In its own words:

“Although individual AAFCO members have the authority to regulate animal feed within their jurisdictions, AAFCO itself has no regulatory authority whatsoever.“

It is a voluntary association of state and federal feed control officials that writes model language states may adopt. Industry participates as non-voting advisors.

“AAFCO approved” does not exist. Again in its own words:

“AAFCO does not approve, certify or otherwise endorse pet foods. There is no AAFCO-approved pet food.”

If a package, a website or a shop assistant tells you a food is AAFCO approved, that is not a strong claim badly phrased. It is a claim about a thing that does not exist.

And the FDA does not approve pet foods before they are sold either. It regulates them as food, requires them to be safe and truthfully labeled, and acts after a problem is identified. It does not review a formula in advance.

What changed in 2024, and where it stands now

The formal arrangement between the two got looser, not tighter.

The FDA–AAFCO Memorandum of Understanding expired on 1 October 2024 and was not renewed. FDA’s letter to stakeholders framed it carefully:

“The relationship between FDA and AAFCO is not ending, but it will be evolving.”

What replaced it is two guidance documents and a consultation process. The Animal Food Ingredient Consultation is described by the FDA itself as

“an interim process providing an additional way for firms to consult with the FDA regarding animal food ingredients and for the FDA to identify potential safety concerns”

A consultation is not an approval, and “interim” is the FDA’s word, not mine. As of the agency’s most recent published status, its evaluation of its pre-market animal food programs is still under way. So the honest description of the current position is: a standard written by a body with no regulatory authority, substantiated by the manufacturer, in a system whose federal coordination agreement lapsed and has not yet been replaced with anything permanent.

Does AAFCO approve or test dog food?

No. AAFCO states plainly that it “does not approve, certify or otherwise endorse pet foods” and that “there is no AAFCO-approved pet food,” and separately that it “has no regulatory authority whatsoever.” It publishes model nutrient profiles and model label language that states may adopt into their own laws. Manufacturers substantiate their own claims against those profiles. Any product or website advertising itself as “AAFCO approved” is describing something that does not exist.

What happens when somebody actually measures

Given all of the above, the obvious question is whether foods carrying the statement actually meet it. Two recent studies went and checked, and the answer is uncomfortable in a specific direction.

German, Melgoza and Torres-Henderson (2025) analyzed 61 adult and senior diets across 25 brands:

“Eighteen percent (18%; 11/61) of products did not meet at least one requirement for AAFCO adult maintenance.”

Then they broke it down by market tier, and the pattern inverted every assumption in the pet aisle.

Seventeen individual failures, all in minerals: potassium below the minimum in five products, calcium above the maximum in six, phosphorus above the maximum in four, and calcium-to-phosphorus ratio above 2:1 in two.

Price did not predict compliance. It inverted it. The cheapest foods in the study were the ones that met the standard.

A second study, So and colleagues (2026), tested 96 commercial dog and cat products against their own printed guaranteed analysis:

“82.3% of products met label-declared guaranteed analysis values for seven mandatory nutrients.”

Which means roughly one in six did not match its own label. In the puppy foods specifically, 22 percent fell below the calcium minimum, 42 percent below phosphorus, and 72 percent had no detectable EPA and DHA at all.

Grade both honestly. These are single studies with modest samples and exploratory designs, and neither names most of what it tested. They do not prove that boutique food is bad or that mass-market food is good. What they do show is that the statement on the bag is a claim, not a measurement, and that the price of the bag tells you nothing about whether the claim holds.

Do dog foods actually meet their AAFCO claims?

Usually, but not always, and price is no guide. In a 2025 analysis of 61 adult and senior diets, 18 percent failed at least one AAFCO adult maintenance requirement — none of 23 mass-market products, but 8 of 20 boutique and premium products. A separate 2026 study found 82.3 percent of 96 products met their own label’s guaranteed analysis, meaning about one in six did not. All the failures in the first study were mineral levels, which are exactly what a formulation calculation is supposed to control.

So is the statement worth anything?

Yes, and this is the part that would be dishonest to leave out.

A complete and balanced food is enormously better than the alternative most people would build. Home-prepared diets fail nutritional analysis at far higher rates than commercial ones, and the failures tend to be larger. A calculated formula against a published profile, made by a company with a nutritionist, is a good floor. It is the reason severe nutritional deficiency in dogs is now rare enough to be notable when it happens.

Nothing in this article is an argument for feeding something without the statement, or for improvising at home without a board-certified veterinary nutritionist. The point is narrower and more useful:

“Complete and balanced” is a floor, not a ranking. Two foods can both carry it and be very different foods.

Every food in that pet aisle carrying the statement has cleared the same bar. The claim cannot tell you which of them is better, because it was never designed to.

The questions that actually separate one food from another

Since the statement cannot rank foods, something has to. These are the questions worth asking a manufacturer, and the answers are not on any label:

1. Do you employ a full-time qualified nutritionist? A board-certified veterinary nutritionist or a PhD in animal nutrition. Ask for the name and the credential.
2. Who formulates your diets, and what are their qualifications?
3. Do you own your manufacturing plant, or use a co-packer? Owning it is not a guarantee, but it changes who is accountable when something goes wrong.
4. What quality control measures do you use, and do you test incoming ingredients or accept a supplier’s paperwork? This one separates more companies than any other.
5. Have you run AAFCO feeding trials on this product, or only formulated to the profile?
6. Can you send me the complete nutrient analysis for this food, on a dry-matter and caloric basis? Not the guaranteed analysis — the typical analysis.
7. What research have you published, and where?

The answers matter less than the answering. A company with a nutritionist on staff can produce a typical analysis in a few days. A company that cannot has told you something the label never would.

Reading a dog food label covers what the panel can and cannot tell you, and how much to feed covers the one dietary factor with better evidence behind it than anything in this article. The dog feeding calculator turns that into a number for your dog and your bag.

What to do this week

1. Find the nutritional adequacy statement on the bag you are feeding right now. Small print, near the feeding directions. If you cannot find it, that is itself an answer.
2. Note which route it used — “formulated to meet,” or “animal feeding tests substantiate.”
3. Note the life stage, and check it matches the dog in front of you.
4. Email the manufacturer question 6 above and see what comes back, and how fast.
5. Stop treating price as evidence. The measured data says it is not, and in one study it ran the other way.
6. Do not change your dog’s food because of this article. If the food carries the statement and the dog is doing well on it, that is a reasonable place to be. Raise anything that worries you at the next routine appointment.

Who benefits from the confusion

There is a specific commercial reason this stays murky, and it is worth naming.

“Complete and balanced” is a floor that almost everything clears, which makes it useless as a marketing weapon. No brand can differentiate on a claim its competitors all carry. So the differentiation moves to the front of the bag, where the words are unregulated — premium, holistic, ancestral, biologically appropriate, human grade, performance — and where nobody has to substantiate anything to anyone.

Meanwhile the two things that would genuinely differentiate a food — a feeding trial, and a nutritionist on staff — are expensive, invisible on the shelf, and impossible to put in a shelf-talker. So the money goes to the adjectives instead.

The regulated sentence is in six-point type near the feeding directions. The unregulated words are three inches tall on the front. That is not an accident, and it tells you where the marketing budget went.

If you are feeding a dog on a fixed income and have been made to feel that the affordable bag must be the worse one — the measured evidence in this article says the opposite at least as often as it says anything. Read the small print, ask the manufacturer one question, and stop paying for the adjectives.

—

Related reading: What working and service dogs actually need to eat is the full picture this article sits inside. How to read a dog food label is the panel-by-panel version. The grain-free question is what happens when an unregulated word on the front outruns the evidence, and how I review things explains why no product is named anywhere in this article.