CanineMatch helps humans find dogs (and other animals) in shelters to be adopted.

The list of questions I would ask the team were grouped into 3 main buckets to be asked when appropriate. I chose these 3 buckets to direct the team towards a design that would be not only efficient, but also impactful.

In case you love seeing progress videos and are a fan of "Slow Television" (wiki), here's a video of me generating questions and categorizing them based on how they overlap and interact.
  • What's wrong? – dealing with the problem itself
    • Is the process of finding dogs in shelters different than finding other animals?
    • Who is involved?
    • What do they go through?
    • What do they struggle with?
    • Why do they struggle?
    • Who else tackles this problem?
  • What does success look like? – helps define the scope
    • What do we hope to achieve?
    • How do we measure success?
    • What is the end game?
    • Will it scale?
    • What is required for an MVP?
  • Is our solution feasible? – helps stay within scope
    • Do we have the resources?
    • What are the limitations?

The following are my assumptions to these questions

Is the process of finding dogs in shelters different than finding other animals?

  • Not significantly. While dogs have specific vaccinations and health dependencies, the overall process of adopting a dog is the generally same as any other animal in an animal shelter. For product scalability, let's move forward thinking about all animals in shelters. We might need to work on a new name.

Who is involved?

  • adoption and rescue shelters hoping someone will adopt a pet
  • prospective pet owners looking to adopt a pet
  • veterinarians that ensure the animal is ready for the shelter and adoption

What do they go through?

  • Shelters:
    • Shelters take in stray or unwanted animals.
    • Depending on the animal's history (if available) as well as local laws, the shelter must provide medical care through veterinarians.
    • A profile must be created that includes both photos and descriptions of the animal in order to attract a potential adopter.
    • They continue to care of the animal while answering phone calls and emails and showing the animal to those interested.
    • Once someone is intent on adopting the animal, the shelter may interview the person to make sure the animal would be a good match and to ensure it would be cared for in an appropriate environment.
    • If there are no other issues, the shelter processes the necessary filing and fees associated with the adoption.
    • Finally, the animal's listing is taken down.
  • People looking to adopt:
    • They begin by browsing listings available on shelter/adoption websites and apps. Many listings contain information on the animal's medical history.
    • Once they find an animal that interests them, they call the shelter to see if the animal is still there and available for adoption.
    • A visit to the shelter provides an opportunity to meet the animal and to see if adopting the animal would be a good fit for the adopter's life and environment.
    • Often times, an interview is involved in order for the shelter to assess if the adoption would be good for both the adopter and the animal being adopted.
    • If all goes well, the new owner pays the adoption fee and proceeds to license their new pet.
  • Veterinarians:
    • Shelters often need services provided by veterinarians for animals brought to them. For example:
      • Check-ups
      • Vaccinations
      • Sick animals
      • Animals that need to be spayed or neutered
    • These veterinarians continue to care for these animals in times of sickness
    • After adoption, veterinarians may also provide care for the animal's lifetime if the owners choose to return to that veterinarian.

What do they struggle with?

  • While there are many parts of the adoption process that are difficult, the hardest part is finding a match between a potential owner and a future pet in a timely manner.
    • The shelter must care for the animal until someone adopts it, which consumes both time, space, and money. More importantly, if the animal is not adopted within a certain time frame, the shelter may ultimately have to euthanize the animal.
    • For the person looking to adopt, it's often hard to gauge how compatible an animal would be without going to the shelter and meeting the animal in person. This also consumes time and money (transportation). Furthermore, this visit does not guarantee a compatible match, which then restarts the process.

Why do they struggle? What makes it difficult?

  • The tools and resources currently used to find animals to adopt lack the ability for the adopter to assess the behavior and personality of the animal. Websites and apps currently use a mix of photos, descriptions, and health history to attract a prospective pet owner.
  • While it is possible for the description to convey what type of personality an animal has, it is time intensive for both the shelter to write the description and the adopter to read the description. Furthermore, the description of the animal's behavior is mostly written from the shelter's perspective, which could be quite different from how a prospective owner might observe the animal in person.

Who else tackles this problem?

  • The vast majority of web services and apps provide a platform to list pets up for adoption. While they've made other aspects of the process easier, like listing an animal and providing more visibility to a wider audience, they haven't addressed how to show an animal's behavior.
  • Outside of the pet adoption world, dating apps have a similar purpose in searching for a compatible match in life. While some dating apps like Tinder primarily focus on looks and therefore suffer the same problem with personality, others services like LoveFlutter have put a heavier focus on machine learning of behavior profiles to better match their users.

What do we hope to achieve?

  • By reducing the amount of time it takes to find a compatible match, we decrease an animal's time spent at the shelter.
  • By decreasing an animal's time spent at the shelter, we decrease the overall cost and labor associated with caring for that animal. This results in more space for shelters to care for more animals.
  • By enabling shelters to care for more animals, we reduce the rates at which animals are euthanized at the shelter.

How do we measure success?

  • A decrease in an animal's time spent at a shelter.
  • A significant decrease in the number of euthanized animals at shelters.

What is the end game?

  • CanineMatch becomes the industry standard for pet adoption, leading to less animals dying at shelters worldwide.

Will it scale?

  • Yes. User's usage is limited by their distance to the shelter. Will scale as more shelters participate.

What is required for an MVP?

  • A shelter should be able to create a profile for each animal they care for.
  • A shelter should be able to convey how an animal behaves without resorting to text descriptors.
  • A shelter should be able to provide contact information.
  • A prospective pet owner should be able to browse profiles of animals in need of adoption within a defined area.
  • A prospective pet owner should be able to easily determine basic personality traits of each animal.
  • A prospective pet owner should be able to contact the shelter to continue the pet adoption process.
  • The major purpose should be to highlight the behavior of each animal and see if it has a positive effect on adoption rates in the real world.

Do we have the resources for a 10 week implementation?

  • I'm assuming, yes.

What are the limitations?

  • For the product, avoid emerging tech and hardware that everyday people might not have access to. For example:
    • VR
    • IoT hardware
  • For MVP, avoid technology with heavy investment and R&D. For example, machine learning.


The following are rough timelines that are not rigid and items can generally move ±1 week depending on how the project goes and depending on what we find through testing and iterations. Most of the items are collaborative across teams, but I wanted to highlight the moments where communication and collaboration are critical to the success of the project.