CEO, Save All The Pets
Posted April 30, 2019 at 05:03 AM Under "Data & Technology"
Saving lives with data science and relocation

I’m starting a nonprofit to save the lives of dogs and cats by finding out where they are statistically more likely to be adopted and then moving them there. A description is below:

With a team of data scientists, I am collecting data about what dog traits make a dog more likely to be adopted in which locations. After I obtain the data, I will analyze it by breed, age of dog, and size, and then relocate dogs to where they’re statistically more likely to be adopted. For example, if terriers are more likely to be adopted in Dallas than Houston, we’ll move a certain number there according to the model. Given that 800,000 dogs can cats per year are killed nationwide in shelters, this has the promise to save many pet lives.

I am using machine learning and data science techniques to find the optimal distribution of dogs to ensure adoption. I’m also using Google image recognition technology to identify the breeds in photographs since they are often mislabeled. Natural Language Processing will be used to determine if there are any trends to uncover in the dog's description on PetFinder/Adopt-a-Pet.

After I get the process in place with dogs, I’ll expand to cats. I'm working with the head of a company called Doobert that has more than 23,000 volunteers who transport animals all over the nation, and I am also exploring partnerships with other organizations.

I’d love to hear suggestions and ideas from this forum about ways to make this even better and ensure success. I’m currently talking to major media outlets and am excited about the prospect to save lives. Also feel free to reach out to me privately if you’d like to contribute. Thanks so much!

  • May 02, 2019 at 04:50 AM

    I love data, so this sounds really interesting to me.

    Maybe you could also consider where these animals are most likely to be euthanized, as some shelters have extremely high kill rates, when deciding where to pull from?

  • May 01, 2019 at 10:13 AM

    I love your idea and was just sharing with our shelter staff yesterday that we need to "go Moneyball with our shelter data/placement stats"!

    I have studied our shelter data a little bit, the types that come and where we are able to place them. Even as an underfunded open admissions shelter in an urban environment, through the incredible work of so many staff/volunteers/donors we are able to place most cats, small dogs (even with medical/behavior issues), and healthy large dogs under ~5 with no behavior issues. Pit bulls / bully breeds, German Shepherds, huskies or Malinois with even minor / simple issues, or seniors, are a very different story. Our rescue partners try to help, but they can't take them anywhere near the rates that they come in our doors each day. Could add to that "solid colored Chihuahuas who are either old, shy/fear biters/or have medical issues". So, what would be super helpful for us to know is where there are rescues who can take these particular breeds, but need to go beyond breed to address: senior dogs; feral cats; dogs with medical issues; dogs with behavior issues such as dog reactivity or food guarding or shy. Also would be good to know which groups can compensate the municipal shelter for some of our costs involved with animal care, vaccines, medical, transportation - particularly the highly sought after animals that are readily placed such as anything fluffy, labs, etc. I know this is common in the east but not in the west. This would help us to save the big dogs that are so much harder to place -it could help us cover their medical/transport costs. The bottom line for us is that the endless stream of pitbulls / bullies, Malinois and shepherds coming through our doors at much faster rates than it seems possible to find placement is overwhelming. We struggle to place the well behaved young healthy ones...we have not found a miracle solution for the ones with any type of issue.

    Another avenue for some of these dogs is private adoption / placement for owned dogs that people can't keep any longer due to the various reasons they end up getting surrendered. Particularly the dogs in the hard to place categories I listed - if we had groups in different locations who could work with families BEFORE they need to surrender, without any judgment (not always easy), it would be a huge win for us.

  • April 30, 2019 at 05:03 PM


  • April 30, 2019 at 07:25 AM

    This sounds like an amazing tool, particularly for those shelters in the North that don't have enough animals and the shelters in the South that have too many. I'll be anxious to see how this project progresses!