Live from the City of Austin open data portal
Earlier attention. Shorter waits. More animals home.
SecondChance helps shelter teams see which animals may need extra support, then turns that insight into a practical outreach plan.
Live Austin feed
open shelter stays
1,048 cat and dog stays with an open intake record and no matched outcome. Not a roster of adoptable animals.
Scores route help toward animals. They must never be used to justify euthanasia, intake refusal, or reduced care.
01Overview
Open shelter stays
stays
Open intake records at the snapshot. Not a roster of adoptable animals.
What this means
Cat and dog intake episodes with no confidently matched recorded outcome as of the snapshot.
- Window:
- As of the snapshot time shown in the data details.
- Sample:
- Wildlife and livestock are excluded. Ambiguous joins are excluded. Records open for more than a year are counted but flagged as possibly stale.
- Use:
- A rough size for the population that may need review.
- Limit:
- Not verified adoption availability and not exact kennel occupancy: Austin publishes intake and outcome events, not a roster.
Median days to adoption
17days
Adopted animals only, stays begun in the last twelve months.
What this means
The middle wait among animals actually adopted in the cohort: about half waited less and half more.
- Window:
- Stays that began in the trailing twelve months.
- Sample:
- Adoptions only. Transfers, returns to owner, euthanasia and open stays are not in the sample.
- Use:
- Comparing observed adoption timelines across periods or groups.
- Limit:
- It is not every animal's expected wait: animals still waiting and animals who left another way are not counted.
Adopted within 30 days
31%of eligible arrivals
Arrivals too recent for a full 30 days are excluded, not counted as failures.
What this means
Eligible arrivals adopted by day 30, divided by arrivals that have had a full 30 days in which to be adopted.
- Window:
- Arrivals in the trailing twelve months, excluding the most recent 30 days.
- Sample:
- Arrivals too recent to have had 30 days are excluded, not counted as failures. Other exits within 30 days count as not adopted.
- Use:
- Historical early-adoption frequency.
- Limit:
- A recent change in practice shows up only after 30 days have elapsed.
Stays longer than 60 days
7%of completed stays
What was observed in completed history, not what the model predicts.
What this means
Completed stays lasting more than 60 days, divided by completed stays.
- Window:
- Stays that began in the trailing twelve months.
- Sample:
- Completed, confidently matched cat and dog stays, whatever the outcome.
- Use:
- How often long stays happened in completed history.
- Limit:
- It is not the fraction of animals currently here who are past 60 days, and it is not a prediction.
02Attention
Needs attention
The five open stays with the highest intake-time long-stay estimates. Each row leads with how long the animal has actually waited.
- Already waiting over 60 daysintake estimate was 71%
Latke18333
Dog · American Pit Bull Terrier
- Already waiting over 60 daysintake estimate was 71%
Leo21487
Dog · American Pit Bull Terrier
- Already waiting over 60 daysintake estimate was 71%
Fiesta24646
Dog · American Pit Bull Terrier
- Already waiting over 60 daysintake estimate was 71%
Truffle24977
Dog · American Pit Bull Terrier
- Already waiting over 60 daysintake estimate was 71%
Damon27222
Dog · American Pit Bull Terrier
03Trends
Intakes and adoptions by month
Cats and dogs, from the current ShelterBuddy feeds. Two event counts, not one cohort.
Intakes are counted by arrival month and adoptions by outcome month. They can involve different animals, so the lines are not a rate; the 30-day adoption rate is a separate cohort figure.
Where the wait is longest
Observed median days to adoption by life stage at intake, last twelve months
- Adult18 daysbased on 1,893 adoptions
- Under a year16 daysbased on 3,065 adoptions
- Senior16 daysbased on 136 adoptions
- Age not recorded12 daysbased on 27 adoptions
The median counts adopted animals only; the adoption share divides adoptions by all completed stays. A short median can sit beside a low share. Life stage is age at intake; the archive study used a different era and metric, so its ranking can differ.
How stays end
Completed cat and dog stays that began in the last twelve months
- Adopted49%
- Transfer Out34%
- Reclaimed9%
- Euthanized3%
- Return to Habitat2%
- Other outcomes2%
Each share is that outcome divided by completed stays that began in the window. Transfer is not adoption; open stays have no outcome yet and are not included.
04Model context
Model context
How it worksA validated model scores every eligible open stay. Its evaluation is summarised here and set out in full on How it works.
Show the evaluation summary
- 0.694
Discrimination (ROC AUC)
How often the model ranks a long stay above a short one, on the held-out block. 0.5 is chance.
- 0.047
Calibration error (ECE)
Average gap between stated and observed rates, in probability points. A model can rank well and still say the wrong number.
- sc-sb-2026-09-02-v1
Model version
Read from the published artifact, not from a constant in the interface.
- 2026-09-02
Trained
The moment training ran against the current ShelterBuddy feeds. It changes only after a documented retrain.
05How to work with it
Read the pattern, review the animal, plan support
Step 1
Read the pattern
The tiles and charts describe what has happened: waits, outcomes and pressure over the last year.
You are here
Step 2
Review the animal
The queue puts intake-time estimates next to actual waits, with the record behind each one.
Open the queueStep 3
Plan support
The planner turns a shortlist into a week of outreach work with the assumptions written down.
Open the planner
Cat and dog stays inferred from the public intake and outcome feeds. This is not a roster of animals available for adoption. Why this distinction matters

