“Complete and balanced” is a floor, not a target

Short answer: “complete and balanced” means a food meets published minimums for the nutrients an animal will visibly fail without. It is a floor against acute deficiency, and it is a real achievement — but it says almost nothing about chronic outcomes over a fifteen-year life. We are inside this system too: we validate against NRC 2006 and quote the same profiles. The argument is not that the floor is wrong. It is that a floor is being read as a target.

What the phrase actually certifies

A nutrient profile is a list of minimums, and for a few nutrients a maximum. Meet them and the food may be labelled complete and balanced for a life stage. The nutrients on the list are, broadly, the ones where deficiency produces a visible disease on a timescale short enough to attribute to the food.

Taurine is the textbook case. Take it out of a cat's diet and within months you get retinal degeneration and dilated cardiomyopathy. That is unambiguous, it is fast, and it is exactly the kind of failure a minimum-based standard is excellent at preventing. The taurine story is also the reason the system exists: it was a real epidemic, and the standard fixed it.

So the framework is good at what it was designed for. Hold on to that, because the rest of this post is a criticism and it is not a criticism of competence.

The essential-nutrient logic, and where it leaves a gap

Wondering how this applies to your own pet? Build their bowl — two minutes, and it gives the exact grams for their weight.

A profile can only regulate what it defines as a nutrient. Carbohydrate is the clean illustration.

A cat can manufacture all the glucose it needs from protein and fat by gluconeogenesis. There is therefore no dietary requirement for carbohydrate, so carbohydrate is not an essential nutrient, so there is no minimum and — until very recently — no declaration and no maximum. The consequence is structural rather than sinister: a manufacturer meeting the adult-cat protein minimum of 26% on a dry-matter basis has satisfied the standard, and what fills the remainder of the bag is not the standard's business.

For scale, Plantinga et al. (2011) put the natural feline prey diet at roughly 2% of energy from carbohydrate. Extruded dry food commonly runs 40–60%.

An honest correction we owe you. The usual next sentence is “and that is why cats get diabetes.” We went to source it and it does not hold — a 2025 meta-analysis found dietary carbohydrate does not increase body fat or fasted insulin and glucose in cats, with energy excess the actual driver. We wrote that up separately. It matters here because the loophole is real and the popular consequence attached to it is not, and conflating the two is how a good argument gets discredited.

Two structural facts worth knowing

Extrusion needs starch. You cannot make a conventional dry kibble without it — the process depends on starch gelatinising to bind and expand the nugget. So the carbohydrate in dry food is not only a cost decision, it is a manufacturing requirement. A carbohydrate cap would not be a reformulation; for the dry category it would be an existential constraint. That is a fair thing to understand about why the industry position is what it is.

Digestibility is not health. The industry correctly cites research showing cats digest cooked starch with better than 90% efficiency. That is true and it answers a narrower question than it appears to. It passes through the gut usefully and it is the right thing to build a diet on are different claims, and only the first has been demonstrated.

The other floors — the ones we care about more

The carbohydrate case is the famous one. The one that matters for a home-cooked bowl is subtler.

A nutrient profile states requirements per unit of energy. That framing is deliberate and correct, and it carries a consequence people miss: a diet that clears its floors at one level of food intake can miss them at a lower one. A working animal eating a large volume gets its micronutrients as a side effect of the calories. A neutered indoor pet eating a third of that volume gets a third of the micronutrients from the same recipe. Same food, same label, different outcome.

This is the failure mode we built for, and it is invisible if you think in recipes rather than in throughput.

Where we stand, said plainly

We validate against NRC 2006, and we quote AAFCO and FEDIAF profiles in our own cost comparisons. We are not outside this system and we do not claim to be. What we do differently is narrow:

We report the fraction of days a bowl clears its floors across a simulated distribution of realistic inputs, rather than checking one nominal recipe once.
We attach a confidence statement to that fraction, because a number without one is not a claim.
We publish the working under an open licence so it can be checked, including by people who would like to find us wrong.

That is a difference of method, not of moral standing. The floors themselves come from NRC 2006, same as everyone's.

Build your pet's bowl →

References

AAFCO. Dog and Cat Food Nutrient Profiles (current edition) — adult cat maintenance crude protein minimum 26% DM; growth and reproduction 30% DM.
National Research Council (NRC). Nutrient Requirements of Dogs and Cats. Washington, DC: National Academies Press; 2006.
Plantinga et al. (2011). Estimation of the dietary nutrient profile of free-roaming feral cats. British Journal of Nutrition 106(S1):S35–S48.
Rankovic et al. (2025). A meta-analysis: dietary carbohydrates do not increase body fat or fasted insulin and glucose in cats. Journal of Animal Science 103:skaf071.