If you're reading this, you probably already think carefully about what your dog eats. You've moved past supermarket food. You pay more than the average because you want the assurance that you're feeding something better.
That choice is rarely just about convenience. You want something better for your dog. You probably also want something you can feel responsible buying. That's why the environmental question matters.
If you've seen the dog food environmental impact headlines this year, you may have felt some relief if you don't feed raw. The picture turns out to be more complicated than the headlines suggested. The study most of the coverage came from, published this January in the Journal of Cleaner Production, makes a finding that applies across every premium dog food format in the UK market, not just raw. The thing driving emissions in premium dog food isn't the format. It's the ingredients. Specifically, prime meat.
This article walks through what the study actually found, what it means for the premium end of the market, and where our recipe sits within it.
What the study actually measured
The research, led by John Harvey at the Universities of Edinburgh and Exeter, looked at 996 dog foods sold by a single UK retailer. It included 526 dry foods, 197 wet foods, and 34 raw foods, plus grain-free and plant-based variants of each. The team built an open-source model that worked out the carbon footprint of each product from its ingredient list and nutritional declaration.
Across all 996 products, emissions varied 65-fold from the lowest-impact food to the highest. The premium formats sat at the higher end. Raw. Wet. Grain-free. The lower-impact end was standard dry kibble, and a smaller number of plant-based products. Two products on the same shelf, even within the same format, can carry wildly different environmental costs depending on what's in them.
The researchers also estimated that producing the ingredients for UK dog food accounts for around 1% of total UK emissions, and that feeding every dog in the world the way the UK does would generate up to the equivalent of around 99% of the emissions from commercial aviation.
Those are uncomfortable numbers. They're worth sitting with.
What's actually driving the difference
The more useful finding sits underneath the headline. The thing driving the variation between products isn't the format. It's the ingredients. Specifically, prime meat. "Prime" here means prime in human food terms, not in canine nutrition terms. The cuts humans want to eat. Chicken breast, beef steak, lamb leg. These cuts carry the highest environmental burden because they're the parts of the animal that command the highest price and the most agricultural land per kilogram of output. When a dog food uses a lot of them, its footprint goes up.
The reason this matters comes down to how environmental costs get allocated across a whole animal. A beef cow doesn't get raised for one product. It produces hundreds of them. The way researchers allocate the emissions of the whole animal across all those products affects how much each product is held responsible for. Older methods often give every kilogram of output the same share of the impact. Newer methods, including the one the study used, allocate based on economic value. The cuts humans pay the most for carry the most burden. The cuts in lower demand, like organs and cuts that include the bone, carry much less. This is closer to how the supply chain actually works. If humans stopped buying chicken breast, the chickens would stop being raised. If humans stopped buying chicken feet, the chickens would still be raised.
Two dog foods made from the same animals can therefore have completely different environmental footprints, regardless of whether they're raw, wet, fresh cooked, or dry. One built around prime cuts carries the burden of cuts humans compete for. One built around organ meats and cuts that include the bone carries much less, because those are the parts of the animal that aren't in demand for human food. The format on the label matters less than the cuts inside.
The point most coverage missed: The 65-fold variation isn't only between formats. It's within them. Two foods in the same format can have completely different footprints depending on the cuts inside.
What "cuts not typically eaten by humans" means
Harvey made the policy implication explicit in the press around the paper's release. The pet food industry should be using meat cuts that aren't in demand for human consumption. That's the recommendation, in plain language. Use the parts of the animal that humans don't compete for.
What this looks like at the ingredient level is concrete. Minced meat that includes the bone, which delivers protein along with the calcium-to-phosphorus ratio a complete recipe needs. Organ meats: liver, heart, kidney, spleen. Tripe. Gizzards. Lean cuts of beef.
These are real foods. Humans eat them all over the world, in different culinary traditions. Steak and kidney pie, menudo, foie gras, scrambled eggs with chicken livers, blood sausage, sweetbreads. None of this is exotic. It's most of what people eat outside the standard British and American supermarket aisle.
Among premium UK dog food brands, though, these ingredients have largely been treated as lower-grade. Most of the brands competing on quality have moved in the opposite direction. They've moved toward prime muscle meat, on the assumption that what's better for humans must also be better for dogs. The next section looks at whether that assumption holds up.
How premium dog food has framed quality
Most premium dog food brands have organised their pitch around a particular signal of quality: "human grade." Some use the related claim "food you could eat yourself." Both work the same way. They use human food as a reference for what makes good dog food.
The framing was always reasonable. If a food is good enough for a human, the thinking goes, surely it's good enough for a dog. Most pet owners would rather feed their dog something that meets human standards than something that doesn't. As a starting point, that's defensible.
The study suggests it doesn't track quite the way the marketing implies. "Human grade," by definition, means cuts that humans want. The cuts humans want are the cuts driving the highest carbon footprint per kilogram in dog food. A recipe built around chicken breast and beef steak carries more environmental cost than a recipe built around heart, liver, and the cuts humans don't compete for. The same label that signals premium quality to a human buyer signals heaviest environmental burden in the paper recommending the opposite approach.
On nutritional grounds, the picture is similar. A meal made of prime muscle meat alone doesn't give a dog a complete diet. Vitamin A comes from liver. Taurine comes from heart. The calcium-to-phosphorus ratio comes from ground bone. Essential fatty acids come from animal fats. Digestive support comes from tripe. Zinc comes from molluscs. These nutrients aren't optional. They're what makes a recipe work for a dog. A bowl of chicken breast, sweet potato, and broccoli might look like a sensible human dinner. As a dog's only food, it would be missing things the dog needs.
What this means for you: The cuts that carry the lowest environmental cost are also the cuts that carry the broadest nutritional profile for a dog. Environmental best practice and canine nutritional best practice point in the same direction. The signal of quality these brands have chosen has been pointing the other way.
If you've felt something was slightly off about the premium dog food market - that prices keep climbing while the food keeps being presented as more and more like a human dinner - you weren't imagining it. Premium dog food brands have converged, one after another, on the same vocabulary. "Prime cuts." "Human grade." "The kind of food you'd eat yourself." These phrases appear across most of the major brands. The pattern has trained a generation of dog owners to read them as a signal of quality, and that learned reading is now what drives the demand. The cost falls in two places. On the dog, fed cuts whose nutrient profile a recipe then has to supplement back to completeness. On the planet, through the emissions Harvey's paper documents. Pet owners pay more for a food whose central promise the research doesn't fully support.
If you're feeding a fresh cooked or human-grade food right now, this isn't on you. The premium end of the market has spent the past decade teaching every buyer that human-grade is the marker of quality. Most of the brands believed it themselves. You did what the brands said was the right choice. The research just suggests the marker these brands chose doesn't fully track to outcomes the way it sounded like it would. Neither to emissions, nor to canine nutrition.
Our food is real food. Every ingredient is something humans eat somewhere in the world, prepared one way or another. The recipe just isn't designed for a person. It's designed for a dog, and dogs need a different mix of nutrients than humans do. More organ meat. More animal fats. Different vitamin sources. A specific calcium-to-phosphorus ratio. The recipe is built around what a dog needs. How it would look on a human plate wasn't part of the brief.
What our recipe is actually made of
The full ingredient list is on every box. The structure is straightforward.
The beef portion includes heart, lean meat, kidney, spleen, green tripe, and liver. The chicken portion includes minced chicken with ground bone, and gizzards. Together they make up around 89% of the recipe by weight. The recipe is built around the kinds of cuts the study points toward: organs, minced chicken with ground bone, green tripe, gizzards, and other cuts humans don't typically compete for in the UK market.
The remaining 11% is the part of the recipe that makes it complete. Whole egg. Fish oil, cod liver oil, and flaxseed oil for omega-3 and omega-6 fatty acids. Sunflower seed meal for linoleic acid. Pumpkin fibre and psyllium husk for digestive support. Kelp meal for iodine. Dehydrated oyster for zinc. Dehydrated mussel for manganese. Nutritional yeast for B-vitamins. These are also real foods, sourced as ingredients rather than as synthetic additives. The recipe doesn't use a vitamin and mineral premix. Every nutrient comes from something a dog can recognise as food. See the full ingredient breakdown for the nutrient-to-source mapping.
What we can't claim, and won't
We don't have a Life Cycle Assessment of Protocol Raw Complete yet. That means we can't put a specific CO2-equivalent number on a kilogram of our food. We can't claim to be the lowest-impact dog food on the UK market because we haven't run the comparison. The study doesn't include Protocol Raw, because it sampled a single retailer and we're not in retail.
What we can say is that the formulation is built around exactly the kind of ingredient mix the paper's own policy recommendation points to. That is a claim about how the recipe is built, not a measurement. We have not run our own environmental comparison. The structural fact stands. The specific number doesn't yet. Running our own LCA is on the list of things to do post-launch, when there's a real production volume to assess and the numbers will mean something. We'll publish what we find.
What we'd support
Two things in the paper are worth supporting plainly. First, the call for clearer ingredient labelling. Dog owners can't make informed choices about environmental impact when most labels describe ingredients in generic terms like "chicken" or "beef meal" rather than specific cuts. A labelling standard that names the actual parts of the animal would help everyone. Second, the broader point that pet food belongs inside the wider food sustainability conversation. It hasn't been, for most of the history of the industry. It should be.
Our own contribution is structural rather than rhetorical. The recipe is what it is. The ingredient list is on every box, and the cuts are named specifically. If the category moves toward better labelling, we already meet the standard.
The reference
Harvey, J., Crowley, S. L., Eory, V., & Alexander, P. (2026). Estimating the Environmental Impact of Dog Foods Marketed in the UK. Journal of Cleaner Production, Volume 540, Article 147277. The paper is open access under Creative Commons. Read it here.
For more on how raw differs from fresh cooked dog food, see Raw vs Cooked Dog Food: What the Evidence Says.