Puppy DNA Test Shows a Surprise Breed? Here's What It Usually Means
Quick Answer: If your puppy DNA test shows a small amount of a breed that's a close relative of your puppy's breed, it's usually the test confusing two cousin breeds, not a sign of a hidden cross.
A puppy DNA test is a fun thing to do with a new dog. Most of the time the report says just what you expected. But sometimes it doesn't.
Say you brought home a purebred Pembroke Welsh Corgi puppy. You send off a DNA kit, and the report comes back 97% Pembroke and 3% Cardigan Welsh Corgi.
That can be upsetting. Did your breeder get it wrong? Is your puppy really a mix? Most of the time, the answer is no. To see why, it helps to know how these tests actually work.
How does a puppy DNA test decide what breed my dog is?
A DNA test doesn't find a tag on your puppy's DNA that says "Pembroke." There's no label like that.
Instead, the test cuts your puppy's DNA into lots of small pieces. Then it compares each piece to a big library of DNA from dogs of known breeds.
For every piece, it asks one question: which breed does this look most like? Add up all those answers, and you get the percentages on your report.
That works really well when breeds are very different. A Pembroke and a Chihuahua don't look much alike on the inside either. But it gets harder when two breeds are close relatives.
Why do close cousin breeds get mixed up?
Think of breeds like a family tree. Some breeds are like distant strangers who live across the globe. Others are like cousins who grew up in the same house.
In 2017, researchers at the National Institutes of Health tested 1,346 dogs from 161 breeds and built a family tree showing how breeds are related. They sorted the breeds into family groups. Scientists call these groups "clades."
The chart above comes from an earlier study (vonHoldt et al., Nature, 2010) that also sorted breeds into family groups. Look at the green herding section. The Pembroke and the Cardigan sit right next to each other.
That's no surprise. Until the 1930s, kennel clubs registered them as one breed. They share a lot of the same DNA.
And this isn’t just a Corgi thing. Lots of breeds have close cousins on the family tree. Labrador Retrievers and Golden Retrievers, for example, sit in the same family group as other retriever breeds. So a Lab’s report might show a small amount of another retriever for the same reason.
So when a piece of Pembroke DNA looks a tiny bit more like the Cardigan dogs in the library, the test calls it Cardigan. The DNA didn't change. The test just made its best guess between two cousins.
There's a second reason. A DNA test is only as good as its library of reference dogs. If a breed has only a small number of reference dogs, or those dogs come from different bloodlines than yours, the test has less to go on.
Embark explains that a registered purebred may not perfectly match its breed's reference dogs. The dog might have an ancestor from a closely related breed used long ago, before the breed's registry closed. Or its bloodline might come from a very different part of the world than the reference dogs.
Both Embark and Wisdom Panel say they have the largest reference database. Neither one publishes how many reference dogs it has for each breed, and that's the number that matters most here.
Why would my puppy show it when her littermates didn't?
Every puppy gets half her DNA from mom and half from dad. But it's not the same half for every puppy. Each one gets a different random mix.
That's why littermates can have different breed percentages even when they share both parents. Your puppy may just have a few more of those "Cardigan-looking" Pembroke pieces than her brothers and sisters. Nothing unusual happened. She just got a slightly different hand of cards.
What percentage is too small to worry about?
Embark doesn't publish a set cutoff. When a piece of DNA is too small or too old to match to one breed, they put it in a category they call "Supermutt" instead of naming a breed.
Many breeders use about 3 to 5% as a loose rule of thumb for a gray zone. It isn’t an official standard or a hard cutoff. But the math behind it helps explain why small amounts are so hard to read.
Parent: about 50%
Grandparent: about 25%
Great-grandparent: about 12.5%
Great-great-grandparent: about 6%
One more generation back: about 3%
Real inheritance is random, so these are averages, not exact numbers. But once you get under 5%, you're talking about an ancestor four or five generations back. At that level, a real distant ancestor and a test guessing between cousins can look exactly the same.
Here's a practical way to think about it:
A small amount of a close cousin breed: most likely the test mixing up related breeds.
A small amount of a breed from a very different family group: worth a closer look. A Pembroke showing 4% of a breed from a far-off branch of the tree is harder to explain as a cousin mix-up.
A larger amount: worth a calm conversation with your breeder and the testing company.
Does this mean my puppy isn't purebred?
No. Embark says this plainly in its own help center. A DNA test can't always rule out a small amount of a closely related breed, and a small percentage of another breed doesn't cancel a dog's purebred status.
Purebred is about registration and pedigree. The DNA test is a separate thing.
What should I ask my breeder?
Here's the most important thing to know: a puppy can only carry DNA her parents had. If both parents were tested and neither one shows the surprise breed, your puppy can't have gotten real DNA of that breed from them.
You probably can't test the parents yourself, but your breeder may already have. A good breeder will be glad you asked. Some good questions:
Were the parents DNA tested, and can I see their breed results?
Which company did you use? Each company uses its own library of reference dogs, so results from two companies may not match exactly.
Have other puppies from these lines shown something similar?
If the parents tested clean, you have a strong answer. Your puppy's small surprise result is almost surely her own breed being read as its closest relative.
What if a parent shows the same breed too? That doesn’t automatically mean a hidden cross either. Some family lines show the same small amount of another breed, generation after generation. That can happen when the test’s reference dogs don’t match that line very well, or when the line goes back to an ancestor from before the breed’s registry closed. A result that stays small and steady through the generations is more likely about how the test reads that family than about a recent cross.
If you still have questions, contact the testing company. Tell them what you know about the parents' results. They've seen this before.
A breeder who health tests and DNA tests is doing the kind of homework you want. For more on what testing can and can't tell you, see Why Health Tested Parents" Doesn't Mean a Guaranteed
Healthy Puppy. If you're choosing a puppy now, Is an F1 Really Healthier? covers another question families often ask.
FAQ
Does a small percentage of another breed mean my puppy isn't purebred? No. Purebred status comes from pedigree and registration. A small DNA result doesn't change that on its own.
Can my puppy show a breed that neither parent has? It can show up on the report, but a puppy can't inherit real DNA her parents don't carry. If both parents tested clean, the result is most likely a mix-up between related breeds.
Why did my puppy get this result when her littermates didn't? Each puppy gets a different random half of each parent's DNA. One puppy can inherit pieces that look more like a cousin breed than her siblings' pieces do.
What percentage should I worry about? There's no official cutoff. Many breeders use about 3 to 5% of a closely related breed as a loose guide, not a hard rule. A larger amount, or a breed from a very different family group, deserves a closer look.
What's a clade? A family group of breeds that share common ancestors. Breeds in the same clade share a lot of DNA, which is why tests sometimes mix them up.
Sources
Parker HG, Dreger DL, Rimbault M, Davis BW, Mullen AB, Carpintero-Ramirez G, Ostrander EA. Genomic analyses reveal the influence of geographic origin, migration, and hybridization on modern dog breed development. Cell Reports. 2017;19(4):697-708. https://doi.org/10.1016/j.celrep.2017.03.079
vonHoldt BM, Pollinger JP, Lohmueller KE, et al. Genome-wide SNP and haplotype analyses reveal a rich history underlying dog domestication. Nature. 2010;464:898-902. https://pmc.ncbi.nlm.nih.gov/articles/PMC3494089/
Embark Veterinary. Can a registered purebred dog appear as mixed breed on a DNA test? https://embarkvet.com/resources/can-a-registered-purebred-dog-appear-as-mixed-breed-on-a-dna-test-2/
Embark Veterinary. Does an "Unresolved" result mean my dog is not purebred? https://help.embarkvet.com/hc/en-us/articles/17644756224155-Does-an-Unresolved-result-mean-my-dog-is-not-purebred
Embark Veterinary. The most accurate dog DNA test. https://embarkvet.com/resources/most-accurate-dog-dna-test/
Embark Veterinary. What is a "Supermutt"? https://embarkvet.com/blogs/resources/what-is-supermutt
Wisdom Panel. Our science. https://www.wisdompanel.com/en-us/our-science
All web sources accessed September 30, 2026.
