top of page
banner.png
Search

Dog DNA Test Wrong Breed? Why Results Show a Breed That Isn't in the Pedigree

Ji Khalsa
Sep 30
7 min read

Quick Answer: When a DNA test shows a small amount of DNA from a breed that's closely related to your dog's real breeds, it's often reading shared family DNA, not necessarily proof of a hidden cross.


Here's a situation that comes up a lot in breeder groups.


Say you have a litter of purebred Pembroke Welsh Corgis. You've DNA tested both parents. You've tested the grandparents and great-grandparents too. Every one of them comes back 100% Pembroke.


The puppies get tested, and all but one come back just like you'd expect. That one puppy shows 3% Cardigan Welsh Corgi.


Parentage is confirmed. She's definitely theirs. It looks like a dog DNA test wrong breed result. But most of the time, the extra breed didn't come from anywhere. To see why, it helps to know how these tests actually work.


How does a DNA test decide what breed my dog is?

A DNA test doesn't find a tag on your dog's DNA that says "Pembroke." There's no label like that.


Instead, the test cuts your dog'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 from each other. A Pembroke and a Chihuahua don't look much alike on the inside either. But it gets harder when two breeds are close relatives, which you can see a bit more clearly on the chart below.


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." You can read the full study, Genomic Analyses Reveal the Influence of Geographic Origin, Migration, and Hybridization on Modern Dog Breed Development (Parker et al., Cell Reports, 2017).



Circular family tree of dog breeds and wolves, grouped by color into types like herding dogs, retrievers, and scent hounds, from vonHoldt et al., Nature, 2010.
Family tree of dog breeds and wolves. Breeds with the same color are closely related, like the herding dogs in green. Source: vonHoldt et al., Nature, 2010.

The chart above (called a "cladogram") 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.


So when a piece of Pembroke DNA happens to look a tiny bit more like the Cardigan samples 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 this happens. A DNA test can only be as good as its library of reference dogs. Each company builds that library from registered purebred dogs of every breed it tests for.


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 for a couple of reasons. The dog might have an ancestor from a closely related breed that was used long ago, before the breed's registry closed to outside dogs. Or the dog's 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 only one puppy in the litter show it?

Every puppy gets half its 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 siblings can have different breed percentages even when they share both parents.


So one puppy can end up with a few of those "Cardigan-looking" Pembroke pieces that her littermates didn't get. Nothing unusual happened in the breeding. She just got a slightly different hand of cards.


So how do I know if it's real?

Here's the most important thing to remember: a puppy can only carry DNA her parents had.


If both parents are tested and neither one shows any Cardigan, the puppy can't have gotten real Cardigan DNA from them. The small amount on her report is almost surely Pembroke DNA being read as its closest relative. This is sometimes referred to as a genetic "artifact."


This is also a good reason to test your breeding dogs, not just your puppies. When you have results for parents and grandparents, a strange puppy result is much easier to explain. If you use DNA testing for parentage, our post on dual-sired litters covers how that works.


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. Embark also notes that 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 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 a rule of thumb that anything under about 3 to 5% is in the gray zone. That isn't an official standard, but the math behind it makes sense. Here's roughly how much DNA each generation back gives your dog:


  • 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 you can see the point. Once you get under 5%, you're talking about an ancestor four or five generations back.


That's the same range where cousin-breed mix-ups happen. 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:


  • Small amount of a close cousin breed, parents tested clean: most likely a mix-up between related breeds.

  • 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.

  • Larger amounts, or the same breed showing up in the parents: that's worth taking seriously and talking to the testing company about.


What should I do about a dog DNA test wrong breed result?

Start with the parents' results. If neither one carries the breed in question, you have a strong case that it's a mix-up.


Then reach out to the testing company. Explain what you have tested in the family.


And keep good records. Knowing what you have and why is part of setting clear breeding goals. If you're newer to breeding and genetics feels overwhelming, start here. You don't need to be a geneticist. You just need to know which questions to ask.


Can I share this with my puppy buyers?

Yes, please do. Save or bookmark this link and share it on your website, your puppy page, and your social media.


If you see these results now and then in your lines, you can add it to your go-home materials too. That may be more likely if your dogs come from bloodlines in another country, since the test's reference dogs may not include many dogs like yours.


Lots of puppy buyers DNA test their new dog. If a surprise breed shows up, a buyer may worry that something is wrong with their puppy or with your program.


Sharing this ahead of time helps buyers understand what a small cousin-breed result means before it ever comes up. And if it does come up, you can send them this post instead of explaining it from scratch.


FAQ

Does a small percentage of another breed mean my dog isn't purebred? No. Purebred status comes from pedigree and registration. A small DNA result doesn't change that on its own.


Can a 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 test clean, the result is most likely a mix-up between related breeds.


Why did only one puppy in the litter get this result? 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 treat under 3 to 5% of a closely related breed as normal background noise. 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


All web sources accessed September 30, 2026.

As featured in National Geographic, The Wall Street Journal, the Functional Breeding Podcast, and The Breeder Tails.

Official Core Curriculum of the Functional Dog Collaborative.

​

​

Amazon #1 bestselling author of three books on responsible dog breeding and ownership.
 

IAABC_web_bw_Accredited.jpg
adpt transp.png
IACP transp.png
dwaa-logo.png
Long ad.png
  • TikTok
  • Facebook
  • Instagram
  • Youtube
© Copyright 2019-2025 Midwoofery® LLC | ji@midwoofery.com

Privacy policy • Affiliate disclosure • Disclaimers

bottom of page