The lack of successful translation of basic research discoveries into safe and effective treatments for chronic pain patients has led to increased scrutiny of the preclinical models used in pain research, particularly for osteoarthritis, where there is a significant disconnect between the animal models used to study the structural versus symptomatic aspects of the disease. Companion dogs offer a unique opportunity to assess osteoarthritis pain in a physiologically relevant 'model' of the disease. Approximately 20% of the canine pet population spontaneously develops osteoarthritis, translating to at least 15 million dogs in the United States alone. As in humans, pathogenesis of canine osteoarthritis involves changes in all tissues of the synovial joint including articular cartilage, subchondral bone, and periosteum. The dominant symptom of osteoarthritis for both humans and dogs is pain, and the current therapeutic goal for both species is management of that pain and associated loss of function. To capture clinically and translationally relevant pain severity and pain impact data in the companion canine osteoarthritis 'model', clinical metrology instruments have been validated. These instruments, which assess changes in spontaneous pain-related behaviours, over extended periods of time, in the dog's home environment, are used to evaluate the efficacy of novel interventions for chronic pain in canine osteoarthritis studies. There is evidence that these results in companion dogs can reliably predict efficacy in humans. Across many classes of compounds in which there have been studies in companion animal chronic pain conditions and the same conditions in humans, the analogous results have been seen. In addition, many of the drugs used to treat pain in people are successfully used off-label to treat pain in dogs as well. If preliminary indications of predictability hold true, companion dogs may be embraced as a missing link in the translation of osteoarthritis treatment from mice to men.