Embark is one of the most comprehensive dog DNA tests available, but its accuracy depends on what you’re measuring. In an independent study published in the Journal of the American Veterinary Medical Association, Embark correctly identified the breed of purebred dogs 83% of the time (10 out of 12 samples). That number may sound lower than expected, but the full picture is more nuanced, especially when you consider the technology behind the test and what “accuracy” means for mixed-breed dogs versus purebreds.
What the Independent Testing Shows
The JAVMA study tested multiple direct-to-consumer canine DNA kits by sending in samples from dogs with known breeds. Embark matched the correct breed in 83% of cases. That’s a strong result, but not perfect. The mismatches don’t necessarily mean the test produced wildly wrong answers. In some cases, closely related breeds share so much DNA that the test picks up the right breed group but assigns the wrong specific breed within it. Think of it like confusing a Labrador Retriever with a Chesapeake Bay Retriever: genetically close, visually distinct.
It’s worth noting that the sample size in this study was small (12 dogs), so a single misidentification shifts the percentage significantly. A larger dataset might show a different number, but this remains one of the few independent, published evaluations of consumer dog DNA tests.
How Embark Analyzes Your Dog’s DNA
Embark scans more than 230,000 genetic markers from each saliva sample, comparing them against a reference database of over 350 breeds. That marker count is significantly higher than what most competitors use, and it’s the same type of technology (called a SNP microarray) used in human genetic testing. More markers generally means finer resolution, which helps distinguish between breeds that share a lot of common ancestry.
The company was co-founded by researchers at Cornell University’s College of Veterinary Medicine, and the data collected from customer samples continues to feed into veterinary genetics research. Each test also screens for more than 270 genetic health conditions, from exercise-induced collapse to degenerative myelopathy, in addition to the breed breakdown.
Why Mixed-Breed Results Are Harder to Verify
Most people buying Embark have mixed-breed dogs, and that’s where accuracy gets tricky to evaluate. There’s no “answer key” for a shelter mutt the way there is for a registered purebred. When Embark tells you your dog is 30% Australian Shepherd and 20% Pit Bull, there’s no practical way to confirm those exact percentages unless you know the dog’s full pedigree.
What you can look for is consistency. If two tests from the same company return the same results for the same dog, the underlying analysis is reliable even if you can’t independently verify every breed. Embark’s large marker count works in its favor here, giving it more data points to work with when untangling complex ancestry.
What “Unresolved” Ancestry Means
Some Embark results include a percentage labeled “unresolved,” which can be confusing. This happens for two reasons. First, every generation of breeding breaks DNA into shorter and shorter segments. After enough generations, those fragments become too small for the algorithm to confidently match to any single breed, so the test labels them unresolved rather than guessing.
Second, your dog may carry genetic patterns that aren’t yet represented in Embark’s reference database. While 350+ breeds is extensive, some breeds have regional or rare genetic diversity that the database hasn’t fully captured. Embark says these unresolved portions may shrink over time as the reference panel grows. A small percentage of unresolved ancestry (under 5-10%) is common and not a sign that the test failed.
Embark vs. Wisdom Panel
Wisdom Panel is Embark’s closest competitor, and the two tests are more similar than different. Wisdom Panel also screens for 350+ breeds and claims 99%+ accuracy for breed identification, though that figure comes from the company’s own internal validation rather than the independent JAVMA study. Wisdom Panel holds laboratory accreditations (ISO 17025, USDA, A2LA) that speak to its quality control processes.
Embark screens a larger number of genetic markers (230,000+) and tests against 400+ breed varieties when you include subtypes. It also offers a more detailed health screening panel. Wisdom Panel tends to cost less. Both tests will give you a reasonable breed breakdown for most dogs. Where you’re more likely to see differences is in dogs with complex ancestry, where higher marker counts give Embark a slight edge in resolution.
Where Accuracy Matters Most
For most dog owners, the breed breakdown is interesting but low-stakes. The health screening is where accuracy carries real weight. A false positive for a serious genetic condition could lead to unnecessary worry or even influence decisions about breeding. A false negative could mean missing a condition your vet should monitor for.
Embark’s health results are generally considered reliable for the specific mutations they test, because identifying a known genetic variant is a more straightforward task than piecing together breed ancestry. The test either detects the mutation or it doesn’t. That said, carrying a mutation doesn’t guarantee your dog will develop the condition. Many genetic health risks are influenced by multiple genes, lifestyle, and environment. Treat a positive result as useful information to share with your vet, not a diagnosis.
If your dog’s breed results include something unexpected, like a breed you’ve never heard of at 8%, consider that small percentages are inherently less certain. The major breed components (anything above 20-25%) are far more reliable than the trace amounts at the bottom of the list.