Data Insight: What We Learned from a Simple Question About Visible IDs and RTH Rates

Animal shelters are overflowing with pets that have become separated from their people. According to our HASS pilot data, as well as Shelter Animals Count (SAC), “stray” animals make up around 50% of 2023 shelter intake across the United States.  It’s not surprising that organizations are looking for answers to the question, “how can we increase return to home (RTH) rates?” 

Here’s an example from one of the HASS pilot organizations: With stray pets accounting for nearly 80% of total intake and very full kennels at LifeLine Animal Project’s DeKalb County shelter in Atlanta, the LifeLine staff teamed up with their HASS Data & Implementation Coordinator to take a deep dive into the data collected when animals enter their shelter and see if it could help with some questions around RTH. We asked what seemed to be a straightforward question around the pets that arrive with some form of identification. “What type of identification has the highest success rate for RTH outcomes?” 

What the HASS team learned was unexpectedly more complicated.

Microchipping has become seen as a gold standard in pet identification, but we all know that microchips do not always equal happy reunions in the shelter lobby. Unregistered chips, out of date owner information, and “dead end” chips are all frustrating roadblocks that shelter staff deal with regularly. Other types of identification have their own shortfalls and no type of identification is perfect, but is there one that has a better chance of reuniting people with their pets if they become lost, once they are taken to a shelter? Can we use this knowledge to reduce the population burden on shelters? That’s what we set out to learn when the shelter began tracking holds and their results, using PetPoint’s “hold” and hold “release reason” fields.

Visions of opportunities danced through our heads as the initial data collection period rolled by: if physical tags show high RTH success rates maybe we can use the data to construct a grant proposal to purchase a tag machine for the shelter! What if rabies tags are a top contender, and if so, is it due to a rock star clinic in the community that does something different? And what will we learn about those frustrating unregistered microchips?

The time had come to see the preliminary results. We grab the hold reports from PetPoint, insert a pivot table into the spreadsheet and BAM! No magic answers. No achievements unlocked. 

The data was . . . cool, but a lot of data was missing. We were not deterred. This was not very surprising to be honest. Data entry can be seen as an administrative burden to staff who have a lobby full of pet owners and animals and staff and volunteers and donations and poop and everything else that makes up a shelter worker’s day. That doesn’t mean we give up.  

Instead, we looked at the data we did have and found that we actually had quite a bit of information, despite the incomplete records. Of a total of 348 stray pets placed on hold because they had some form of identification, 234 animals (67%) had both the hold reason and hold release reason recorded. When we examined the relationship between the type of identification and the hold release reason for these animals we found that the highest RTH rate belonged to animals coming into the shelter with “possible owner” listed as the hold reason (35%). Next in RTH success was “microchip” (31%) as the hold reason. Surprisingly, the hold reason data showed that very few animals (8 in total) entered the shelter with physical identification such as an ID or rabies tag, and only one of them was reclaimed. 

Huh. “Possible Owner”? What does that even mean?! According to LifeLine staff, this is how they flag animals that arrive with neither a microchip nor physical ID, but do arrive with some piece of the identification puzzle, and it’s up to staff to track down the missing pieces. They have some indication that the animal has an owner but are not yet sure. These are the situations where a finder has a general idea of where the pet lives but doesn’t know the exact address. Or maybe a possible match was made with a lost pet report. Any type of lead that isn’t directly tied to another, more appropriate hold reason and can’t be definitively linked to a person is caught in this broad category of “Possible Owner.”Any animal with a possible lead to an owner offers the best chance of getting a pet back to its home, so this looked like the best opportunity for raising the shelter’s RTH rate and moving pets out of an overcrowded shelter. 

After reviewing the findings with LifeLine staff we came to the conclusion that we wanted to continue our quest to collect and analyze data to answer our original question, but with some adjustments informed by what we learned in our preliminary analysis—the most critical of which is the need for consistent data entry, the foundation of our collaborative project. 

The other adjustment HASS recommended was to add a side quest: learn more about the population of animals reunited with their owners after being placed on hold for “possible owner.” We want to know more about these situations to determine if there is insight as to why the intake team is so successful in finding these animals’ owners, even without a microchip or physical ID. 

Like so many questions we ask of data, this one has led us to more questions. We started with a fairly simple analysis of the relationship between types of identification and RTH outcomes, and ultimately discovered a very different lesson and a surprise:

1. We learned all over again that data can be pretty powerful in offering solutions to overcrowded shelters . . . but only if you take the time to record it.
2. Data can surprise us and break down our most simple assumptions. In this case we looked at what type of ID results in the best RTH rates. Instead we found that pets with no ID but with some indication of an owner have the best RTH rates. 

Rather than view this data project as failure, we embrace the opportunity to look closer at interesting findings like “possible owner,” and to shine a light on a data entry process that needs some TLC. We found the next step in our journey toward learning more about what happens when pets enter the shelter with identification, and we found what we think is actually a pretty magical answer.

Want to try finding more about RTH rates at your shelter? Below are steps specific to PetPoint, but you can do this project using other software systems by finding reports specific to “holds” as your starting point.

1. Learn how and where your staff inputs hold related data in PetPoint. They should be recording this in the animal’s record under the tab “Holds.”
   • If hold data is not entered in this tab you will not have data available to view in the report needed for this project. Stop here and adjust the process to capture hold and hold release data in the correct location. 
2. Confirm that holds are released using “hold release,” and that a reason for the release is captured every time.
   • Pro tips:
     • PetPoint will not let you outcome an animal that is on hold, holds must be released prior to outcoming a record. 
     • Hold release information must be captured accurately at the time of entry, this data CANNOT be edited or back dated later
3. Run the report “Holding: History” under the Animals section of reports. Leave criteria as is, just change the data parameters to see the last 6 months- 1 year’s worth of data. 
4. Export the report to a .CSV file and open in Excel or comparable software
   • You’ll want to make some adjustments to the header row to ensure it aligns with the appropriate columns and has the appropriate titles, as the file may contain placeholder text instead of actual field titles
5. Once the sheet has an accurate header row you can insert a pivot table using “Hold Reason” and “Hold Release Reason” as your X and Y axes, and “Animal ID” counts as your value
6. Voila! You should have a table that shows you the number of animals placed on hold along with their hold release reason. 
7. Do a gut check with your intake team: are these results in line with their experience? If not, why and what could be the reason for the difference in their experience and what the data shows?
8. Analyze your findings! What did you learn about animals placed on hold in your shelter’s care? Can you identify areas of opportunity to increase RTH rates for these animals?
9. Determine next steps:
   • Do any hold reasons or hold release reasons need to be added or removed in order to better capture this information moving forward?
   • Do SOPs need to be updated?
   • If data entry is missing or inaccurate, is staff training needed? 
   • Consider monitoring this report on a regular basis to catch errors in a timely manner and provide feedback to staff
10. Share your findings with your intake team, leadership and key stakeholders. 
11. Revisit on a regular basis to see trends, opportunities for improvement and data entry issues

The HASS data and research team would love to hear from you if you have questions, ideas or would like to share your experience looking at hold data! 

Email HASS at hass-project@americanpetsalive.org.