Summary

Most of what sells a bag of dog food is on the front. Almost everything that matters is on the back. This guide covers dry and semi-moist foods in kibble form, which includes extruded kibble, cold-pressed bites, baked and air-dried products. We are leaving wet and canned food out on purpose, because their high water content needs a different reading.

Here is the whole method in a paragraph. The first six ingredients make up most of the food, so that is where you spend your attention. Named ingredients like fresh chicken, salmon or brown rice tell you more than vague categories like meat and animal derivatives or cereals. Watch for ingredient splitting, where one grain is broken into several entries to push meat up the list. Carbohydrate is rarely required on European labels, so if it is missing you can work it out in ten seconds. To compare two foods fairly, convert the numbers to a dry matter basis. And once you have read the ingredients, read the cooking method, because heat decides how much of that protein your dog can actually use.

Two myths to drop before we start. Grain-free is not automatically better; it usually just swaps rice or maize for potato, peas or lentils, and the total starch can be just as high. And you do not need fifty ingredients. Six good ones, in the right order, beat forty fillers.

Why this guide skips wet food

Wet and canned foods are usually 75 to 80% water. That water sits near the top of the ingredient list and makes the meat look dominant, even when the dry content tells a different story. Comparing a pate to a kibble on label numbers alone is misleading unless you strip the water out first.

To keep things clean we stay with dry and semi-moist foods in kibble form: extruded kibble, cold-pressed bites, baked and air-dried. The reading method below works across all of them, and the only number that shifts between them is moisture, which we will handle directly.

How a European pet food label actually works

Two rules do most of the work. First, ingredients are listed in descending order by weight, measured before cooking. The ingredient at the top was the heaviest in the mix. Second, a brand can declare ingredients in one of two ways.

It can name each individual material (dehydrated chicken, maize, chicken fat, beet pulp), which is called an open declaration. Or it can group them into broad categories (meat and animal derivatives, cereals, oils and fats), which is called a closed declaration. Both are legal under Regulation (EC) No 767/2009 and the FEDIAF labelling code.[1][2]

The category method is not illegal and not always low quality, but it is far less transparent. A line like meat and animal derivatives can be different species and different parts from one batch to the next. When a brand names its ingredients it gives up that flexibility and lets you check its work. That choice alone tells you something about how much it wants you to see.

There is a second clue, on the front of the bag. Under FEDIAF rules, flavour means less than 4% of that ingredient, with chicken means at least 4%, rich in chicken means at least 14%, and a chicken dinner or menu means at least 26%.[3] A large photo of a chicken breast beside the words with chicken can legally mean 4% chicken. Read the wording, not the picture.

Start with the first six ingredients

Because ingredients are ranked by weight, the first few make up the bulk of the recipe. In most dry foods the first five or six ingredients are the food. Everything after that is usually present in small or trace amounts. A long tail of forty herbs and botanicals at the bottom looks impressive and changes very little.

So read the first six closely and skim the rest. Ask three questions. Is there a named animal protein at or near the top? How many of the first six are animal versus plant? And is the same basic ingredient, say rice, appearing more than once under different names?

A quick example. Compare these two openings, which is all you usually need.

Food A: Maize, meat and animal derivatives (min. 4% chicken), wheat, animal fat, maize gluten, rice.

Food B: Fresh chicken, fresh pork, hydrolysed rice, fresh fish, beet pulp, brewer's yeast.

Food A leads with maize and a vague meat category, then a second cereal. The real animal content is hard to pin down. Food B names three animal ingredients in its first four and uses one clearly stated carbohydrate. You learned more from a single line of Food B than Food A will tell you on the whole bag.

Good names versus vague names

Once you are looking at the right part of the label, the wording of each ingredient does the rest. Specific, named ingredients are almost always a better sign than broad categories. Here is how the common pairs compare.

Ingredient splitting and ingredient grouping

These are the two tricks that bend the ingredient order without breaking any rule.

Ingredient splitting takes one ingredient and lists it as several smaller ones. Corn can appear as ground corn, corn gluten meal and corn bran. Rice can appear as rice, rice flour and rice bran. Each entry weighs less than the combined total, so each sits lower on the list, and the meat floats up above them. Read together, those three rice entries might easily outweigh the chicken at the top. If you see two or three versions of the same grain near the top, add them back together in your head.[1]

Ingredient grouping is the opposite move. Instead of naming three cheap plant proteins, a brand folds them into one tidy category such as vegetable protein extracts, or hides several species under meat and animal derivatives. Grouping makes a weak part of the recipe look like a single harmless line.

Simple test: a good label makes it easy to add up the animal content and the starch content. A label built to make that hard is usually built that way for a reason.

Total animal content still matters

Once you can see past splitting and grouping, add up the animal ingredients in the first six. This is the clearest single signal of how meat-rich a food really is. A food that opens with three named animal ingredients before its first carbohydrate is in a different class from one that opens with two cereals and a meat category.

One fair caution. Fresh meat is roughly 70% to 75% water, so a fresh ingredient listed first loses most of its weight during cooking. That does not make it bad, it makes it honest. The point is to look at how many animal entries there are and where they sit, not to assume the top ingredient is the largest part of the finished bite.

Meat meal versus fresh meat

This is where many guides oversimplify. Meat meal is not the enemy. It is rendered meat with the water and most of the fat cooked out, which leaves a concentrated protein, often around 60% protein by weight. A named meal such as chicken meal can be a genuinely good ingredient.

The catch is consistency. The digestibility of animal by-product meals varies widely depending on the source and how hot they were rendered, and studies that measured protein digestibility of these meals for dogs found a wide spread between samples.[13] Generic names like meat meal or poultry meal hide that variation completely, so you are trusting the factory, not the label.

Fresh, named meat avoids that second high-heat cook and keeps protein quality and amino acid availability high. The honest summary is short. A named meal is fine, an unnamed meal is a gamble, and fresh named meat is the safest signal of quality. Where you see fresh meat high on the list and no mystery meals, that is a good sign.

The carbohydrate question, and why it is usually missing

Here is something most owners never notice. European labels are not required to declare carbohydrate. The mandatory analytical constituents are protein, fat (crude oils and fats), crude fibre and crude ash, plus moisture when it is above 14%.[1][2] Carbohydrate, measured as nitrogen-free extract or NFE, is optional. Many high-starch foods simply leave it off, and that silence is itself a clue.

If it is missing, you can calculate it in about ten seconds:

NFE (carbohydrate) = 100  −  moisture  −  protein  −  fat  −  crude fibre  −  crude ash

If ash is not listed, use about 7% for a dry food as a working estimate.[10] So a kibble showing 22% protein, 10% fat, 8% moisture, 3% fibre and 7% ash is carrying roughly 50% carbohydrate, even though the bag never says so. A brand that prints its NFE is doing you a favour and signalling it has nothing to hide.

Compare two foods fairly: the dry matter basis

You cannot compare two foods on their bag numbers if they hold different amounts of water. An extruded kibble might be 8% moisture, while a cold-pressed or semi-moist food might be 16 to 18%. The wetter food looks lower in protein only because water is diluting the%age. Strip the water out of both and compare what is left. That is the dry matter basis, and it is the single most useful piece of maths in this guide.

Nutrient on dry matter  =  ( nutrient as printed  ÷  ( 100  −  moisture ) )  ×  100

An example that flips the result. Brand X kibble shows 26% protein at 8% moisture. Brand Y cold-pressed shows 25% protein at 17% moisture. On the shelf, X looks higher. Convert both:

Brand X:  26 ÷ 92 × 100 = 28.3% protein.

Brand Y:  25 ÷ 83 × 100 = 30.1% protein.

Once the water is gone, Brand Y is the higher-protein food, even though it advertised a lower number. Do this for protein and fat before you trust any comparison between two foods. It takes a minute and it routinely changes the answer.

Grain-free is a marketing word, not a health upgrade

Grain-free does not mean carbohydrate-free or even low-carb. It means the starch comes from something other than cereal grains, usually potato, sweet potato, peas, lentils or chickpeas. A grain-free food can easily carry as much starch as a rice-based one, sometimes more. What matters for your dog is the total carbohydrate load and the%age of starchy ingredients, not whether that starch was technically a grain.

There is also a safety thread worth knowing. Since 2018 the US FDA has investigated a link between certain grain-free diets and dilated cardiomyopathy, a serious heart condition. Among the cases reported, more than 90% of the diets were grain-free, 93% contained peas or lentils, and about 42% contained potatoes or sweet potatoes.[7] Further data showed that the link between grain-free diets and DCM is inconclusive and the FDA themselves admitted in a public statement that there's insufficient data to establish causality.[8][9]

None of this makes grains essential or pulses poison. It simply means the words grain-free on the front of a bag tell you nothing useful about quality. Read the starch sources and the carbohydrate total instead, and treat grain-free as a flavour of recipe, not a grade.

Do not stop at the ingredients: read the cooking method

Two foods with the same ingredient list can deliver very different nutrition, because heat changes protein. The main mechanism is the Maillard reaction, the same browning chemistry that crisps a steak. During high-heat processing, lysine, an essential amino acid, binds to sugars and becomes unavailable to the dog, even though it still counts toward the protein figure on the label.[5] In some commercial dry foods, researchers found that only about 38% of the lysine was still chemically available.[6] The protein%age looked fine. The usable protein did not.

That is why the cooking method belongs in your evaluation, immediately after the ingredient list. Here is how the common methods compare.

The gap is measurable, not theoretical. In a University of Illinois study, an extruded kibble had significantly lower dry matter, protein, fat and energy digestibility than every fresh diet tested. The kibble delivered roughly 79 to 81% of its gross energy as usable energy, against about 84 to 87% for the fresh diets.[4] Freeze-dried and gently cooked foods tend to preserve amino acid digestibility better than high-heat extrusion.[14]

One honest caveat about cold pressing. Lower heat protects protein, but it also cooks starch less thoroughly. A cold-pressed food stays digestible only if its starch was pre-cooked or hydrolysed first, otherwise the carbohydrate can pass through poorly.[11][12] So cold-pressed is a real plus for protein, as long as there is a sensibly prepared carbohydrate behind it.

You do not need fifty ingredients

A long ingredient list is often a marketing decision, not a nutritional one. The first six ingredients are the food. The single blueberry, the sprig of rosemary and the pinch of twelve botanicals at the very bottom are usually there in fractions of a%, sometimes for the label more than the dog. A handful of high-quality, named ingredients in the right order beats a wall of forty entries designed to impress. Judge the top of the list, not its length.

A 7-step label check, under two minutes

1. Read only the first six ingredients. They are most of the food.

2. Look for a named animal protein in position one, such as fresh chicken, chicken meal or salmon, not meat and animal derivatives.

3. Spot ingredient splitting: the same grain under several names near the top. Add them back together.

4. Add up the animal ingredients in the first six. More real animal content is better.

5. Find the carbohydrate (NFE). If it is hidden, calculate it: 100 minus moisture, protein, fat, fibre and ash.

6. Convert protein and fat to a dry matter basis before comparing two foods.

7. Check the cooking method. Gentle methods protect protein; extrusion and baking cook hottest.

Putting it to work: a worked example

Theory fades fast, so let us run the whole checklist on one real European dry food, our own Ethelia Healthy Dogs[15] It is a useful teaching example precisely because it ticks most of the boxes above. Apply the same test to any food you are weighing up.

Its first six ingredients are fresh chicken (30%), fresh pork (25%), hydrolysed rice (25%), fresh fish (10%), beet pulp (7.5%) and brewer's yeast. Three of the first four are named animal ingredients, the carbohydrate is one clearly named source rather than the same grain split three ways, and the full list runs to about fifteen real ingredients rather than fifty. Across the recipe, about 65% by raw weight is fresh animal ingredients.[15]

On the cooking side it is cold-pressed and slow-cooked below 60 °C rather than extruded, which protects protein from the Maillard losses described above, and the rice is hydrolysed, which directly answers the starch-digestibility caveat that catches out many cold-pressed foods. It also prints its carbohydrate figure, NFE 31.5%, for better transparency although it not required.[15]

Run the dry matter maths from earlier and its 26% protein at 16% moisture becomes about 31% on a dry matter basis, with fat near 17.9%. The table below scores it against every practice in this guide.

What about value, since a gentler process costs more? Healthy Dogs is listed currently in Greece at 65,00€ for 12 kg, near 5.42€ per kg, which is premium territory. Independently tested in in-vitro study to be at least 25% easier to digest than regular kibble.[15] The label gives you what you need to judge whether the higher named-meat content and the cold-press process are worth that step up from a commodity extruded kibble. That is the whole point of learning to read it.

No food is perfect, and you should still match any food to your own dog. Even here the label could go one step further and name the fish species rather than listing fresh fish. But as a worked example it shows what a transparent label looks like: named animal ingredients up front, one honest carbohydrate, the carb number on show, a short list and a gentle cooking method. Once you can spot those five things, you can read any European dry food in under two minutes.

Sources

Scientific and regulatory references are listed below. Primary literature, regulators and board-certified veterinary nutrition sources were prioritised over uncredentialed commentary.

1. European Parliament and Council. Regulation (EC) No 767/2009 on the placing on the market and use of feed (consolidated). eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:02009R0767-20180101

2. FEDIAF. Code of Good Labelling Practice for Pet Food (2019). europeanpetfood.org/wp-content/uploads/2022/02/FEDIAF_labeling_code_2019_onlineOctober2019.pdf

3. FEDIAF. Pet Food Labelling: A Guide for Customers (2018), thresholds for with, rich in and dinner claims. europeanpetfood.org/wp-content/uploads/2022/02/layman_s_guide_FEDIAF_Labelling_Code_October_2018.pdf

4. de Godoy MRC, Swanson KS, et al. Apparent total tract nutrient digestibility and metabolizable energy in commercial fresh and extruded dry kibble dog foods. University of Illinois. www.ncbi.nlm.nih.gov/pmc/articles/PMC8279163/

5. van Rooijen C, et al. The Maillard reaction and pet food processing: effects on nutritive value and pet health. Nutrition Research Reviews. www.cambridge.org/core/journals/nutrition-research-reviews/article/maillard-reaction-and-pet-food-processing-effects-on-nutritive-value-and-pet-health/E085D3648D6A209003AB4D0DB72DB8B2

6. van Rooijen C, et al. Reactive lysine content in commercially available pet foods. www.ncbi.nlm.nih.gov/pmc/articles/PMC4473178/

7. U.S. Food and Drug Administration. Questions and Answers: FDA's Work on Potential Causes of Non-Hereditary DCM in Dogs. www.fda.gov/animal-veterinary/animal-health-literacy/questions-answers-fdas-work-potential-causes-non-hereditary-dcm-dogs

8. Freeman LM, Linder DE, et al. Diet-associated dilated cardiomyopathy: the cause is not yet known but it hasn't gone away. Tufts Petfoodology. sites.tufts.edu/petfoodology/2023/02/07/diet-associated-dilated-cardiomyopathy-the-cause-is-not-yet-known-but-it-hasnt-gone-away/

9. Role of Diet as a Predisposing Factor for Dilated Cardiomyopathy in Dogs: A Narrative Review. Veterinary Sciences (MDPI), 2025. www.mdpi.com/2306-7381/12/11/1106

10. Effect of crude fiber and total dietary fiber on calculated nitrogen-free extract and metabolizable energy of dog foods. American Journal of Veterinary Research, 2021. avmajournals.avma.org/view/journals/ajvr/82/10/ajvr.82.10.787.xml

11. Beynen AC. Pressed dog food (2020). Review of cold pressing versus extrusion and starch gelatinisation. www.researchgate.net/publication/340429073_Beynen_AC_2020_Pressed_dog_food

12. An Assessment of Starch Content and Gelatinization in Traditional and Non-Traditional Dog Food Formulations. pmc.ncbi.nlm.nih.gov/articles/PMC9739134/

13. Prediction of crude protein digestibility of animal by-product meals for dogs by the protein solubility in pepsin method. www.ncbi.nlm.nih.gov/pmc/articles/PMC4473152/

14. Apparent total tract nutrient digestibility of frozen raw, freeze-dried raw, fresh, and extruded dog foods. Translational Animal Science (Oxford), 2024. academic.oup.com/tas/article/doi/10.1093/tas/txae163/7908844

15. Ethelia Petfoods (ARGOS Petfoods). Healthy Dogs recipe specification and product data. ethelia.com