In September we re-scored all 38 products we were carrying against our own standard. Seventeen came off the shelf. That is 45 percent of the catalogue, it includes the single most expensive item we sold, and one of the seventeen came off because of a mistake we made rather than anything a supplier did.
Every one of them is published, with its reason, on What We Pulled. This is why we did it and what we learned.
Why a standard nobody fails is not a standard
We built this company on a written test. Five weighted checks: does the panel support the package, is there evidence for what it promises, what does the recall record say, do we know who made it and where, and does the company behave the way it advertises. Products are graded A to D and we carry B and above.
The trouble with publishing a test like that is that it is very easy to publish and very easy to never run. Plenty of shops have a page describing their values. Almost none of them will tell you what those values cost, because for most of them the answer is nothing.
So we ran ours properly, on everything, at the same time, and we wrote down what happened. If the number had come back zero we would have had to admit the test was decorative.
Most failures were not bad food
This surprised us and it is the most useful thing in this post if you buy dog food anywhere. Very little of what we pulled was unsafe or badly made. The failures fell into four patterns, and all four are things you can check yourself in about a minute.
1. The front of the bag disagreed with the back
Four products. Two lines of treats sold as grain free whose ingredient panels lead with oat flour. A jerky we sold as single ingredient chicken whose actual panel runs to six ingredients including glycerin and salt. A topper described as low fat directly above a guaranteed analysis reading 39 percent crude fat.
None of that makes the food harmful. It makes the label untrue, and the label is the entire product as far as we are concerned. If you only do one thing differently after reading this, turn the bag over and check whether the panel says what the front promised.
2. Sold as a meal without the sentence that makes it a meal
Five products. A complete diet is supposed to carry an AAFCO nutritional adequacy statement, which is one line of small print confirming the food is balanced for a life stage. It is the difference between dinner and a topping.
Five things we were selling as meals did not have it anywhere, including on the manufacturer's own site. We are not saying those are bad foods. We are saying a bag claiming to be a whole diet has exactly one job on its label and we could not confirm it had been done. That is a document problem, and a document fixes it.
3. The promise outran the evidence
Food does not need a clinical trial to be food. A supplement that promises a specific outcome gets held to what the research actually found, and this is where it got uncomfortable for us.
We were selling a glucosamine joint supplement for $49.99. A 2022 systematic review pooled 72 trials of glucosamine and chondroitin in dogs and cats and concluded they should no longer be recommended for pain management in osteoarthritis. The effect size favoured the placebo groups. Merck's veterinary manual says the same thing.
We knew this. We had published an article saying it while the product sat in the shop. Writing the honest article and keeping the product is worse than never writing the article, because it turns your own research into marketing for something you have already disproved.
4. We could not tell you how to use it
A fish oil with no EPA or DHA figure on the bottle. A skin and coat supplement containing vitamin A and vitamin D that lists an amount for neither, which we were cross selling next to another product that also contains both. Omega 3s are dosed on their EPA and DHA content. Without those numbers there is no dose, just oil and optimism.
The one we pulled on safety alone
Deer antler chews. This one failed nothing on paper.
Cornell's Riney Canine Health Center names antlers twice as a cause of fractured teeth. Merck's veterinary manual lists them among the hard objects that break the carnassial, the big chewing tooth at the back, which is one of the more expensive teeth in a dog's mouth to lose. No antler product has ever held a Veterinary Oral Health Council seal.
We knew that too. It is written in our own article ranking the ways to clean a dog's teeth, where antlers come last of seven. Our front door is a toothbrush. We cannot also sell the thing that breaks teeth and expect anyone to take the toothbrush seriously.
The one that was ours
The toothpaste in our own dental kit was described as enzymatic. On the product page, in the page title, in the image captions, in two of these articles.
We sat down to read the panel properly: sorbitol, hydrated silica, water, glycerin, trisodium citrate, flavour, potassium benzoate, citric acid, disodium EDTA, sodium carboxymethyl cellulose, potassium sorbate, zinc chloride. There is no enzyme in it.
It is still a good paste. Zinc and citrate salts slow how fast plaque hardens, the silica is a mild abrasive, and none of that needed us to invent a mechanism that was not there. While we were at it we also corrected something we had been letting people assume: no item in that kit carries a VOHC seal. The bluestem product with a seal is the water additive, which is not in the box.
Nobody caught this but us, and nobody would have. It would have taken ten minutes to quietly change the word and move on. We are writing it down instead, because a standard you only point at other people is not a standard.
What it cost
Our supplement shelf is empty. Every supplement we carried failed, so there are none, and we are not going to fill the gap with something that would fail next month. We are looking for a fish oil that states its EPA and DHA per serving, because omega 3s are the one supplement category with genuinely good evidence in dogs. When we find one that passes, it goes up. Until then that part of the shop stays empty and the menu link is gone.
The pulled products were also, predictably, some of the better selling ones. Supplements carry the margin in this business. That is not an argument for keeping them.
What happens to the seventeen
They came off the same day. Then we emailed each maker with the specific check that failed and the exact document or correction that would change the answer, because most of these are paperwork rather than food. An AAFCO statement brings five of them straight back. A corrected fat figure brings another. A grain free claim that gets dropped from a package fixes two more.
If they send it, we re-score and relist, and we will update that page when we do. We would rather a supplier fix something than lose them. We would just rather lose them than hand you a bag we cannot explain.
And if you bought any of the seventeen from us, email hello@truvapets.com and we will refund it. No reason needed, nothing to send back.
Do this yourself, on any bag
You do not need us for most of this. Turn the bag over. Check that the first ingredient is an animal with a name on it. Find the AAFCO line and read which life stage it covers. Check that any claim on the front appears somewhere on the back. If the product promises a health outcome, ask what study says so. If you cannot answer those in a minute standing in the aisle, that is information too.
Or send us the bag and we will do it and show you the working, free, whether or not we sell it. That offer is real and a person answers.
Sources
- Barbeau-Grégoire et al., "A 2022 Systematic Review and Meta-Analysis of Enriched Therapeutic Diets and Nutraceuticals in Canine and Feline Osteoarthritis," International Journal of Molecular Sciences, 2022. mdpi.com
- Merck Veterinary Manual, "Osteoarthritis in Dogs." merckvetmanual.com
- Cornell University Riney Canine Health Center, "Dental disease and home dental care." vet.cornell.edu
- Veterinary Oral Health Council, accepted products for dogs. vohc.org
- American Veterinary Medical Association, "Pet dental care." avma.org
- AAFCO, "Understanding Pet Food: Nutritional Adequacy Statement." aafco.org