The adoption site will show you a dog. It will not tell you the dog arrived four hundred days ago. Austin publishes that number, just never next to the animal it belongs to.
One frame per dog, ordered by how long it has been waiting. The day count carries that wait: the longer a dog has been here, the more it has burned into the plate. Photographs come from Austin's own adoption listings, so clicking a dog opens the page where you could actually adopt it.
Austin's adoption site does not show any of this. Its listings carry a dog's age but no intake date, no length of stay, and no way to sort by either, so a person browsing it cannot tell that the dog they are looking at arrived last spring. The city publishes that number in its open data and prints it nowhere near the place people adopt. This page is the two halves put back together.
It is the most repeated claim in animal sheltering. I checked it against every completed dog stay Austin has published since 2013: — of them, — ending in adoption. Here is the median wait by coat color.
Black dogs leave faster than white, brown or chocolate. But look what happens when you split each color by whether the dog is a bully-type breed (pit bull, Staffordshire, American bulldog).
Every bully-type coat lands between —. Every other coat lands between —. The two bands never touch: the fastest bully-type coat is still — slower than the slowest coat outside the group.
Black only looked fast because black dogs are the least likely to be pit bulls: — of them, against — of the blue ones.
Median days to adoption by breed, for every breed with at least 400 adoptions on record.
Austin Animal Center publishes intakes and outcomes as open data on
Socrata, no API key required. There are two eras with two different
schemas: an archive frozen at 2025-05-05 (wter-evkm,
9t4d-g238) and the current feeds that replaced it
(pyqf-r2dc, gsvs-ypi7), which refresh hourly.
This page uses both.
The live outcome feed ships a days_in_shelter column. The
archive does not, so each archived outcome is paired with the most recent
intake for that animal that precedes it. Dogs cycle through the
shelter more than once, and matching on animal ID alone would charge a
repeat stray's entire history to a single stay. —
outcomes could not be paired to a prior intake and were dropped rather
than guessed at.
A dog with an intake row and no outcome row. That is not quite the same as being in the building: some are in foster care, and the outcome feed lags. Animals recorded dead on arrival never receive an outcome row either, so — of them were excluded, because without that filter they sit on this page looking like they are waiting for a home.
Length of stay is heavily right-skewed: a small number of dogs stay for years and drag every average upward. Every headline number here is a median. Color cells in the controlled comparison need at least 250 adoptions on each side of the split to appear, because at a lower floor a 35-dog cell was setting the reported spread.
This is one municipal shelter, and Austin is a no-kill city with an unusually high live-release rate, so these waits are likely shorter than the national picture rather than longer. Nothing here is causal. Breed labels are staff visual assessments, and Olson et al. (2015) found staff labelled 52% of a sample pit bull-type where DNA put it at 21% (The Veterinary Journal 206: 197-202), so "bully-type" means what Austin wrote down, which is also what an adopter reads.
The open data has no images in it. The photographs here come from Austin Animal Center's own adoption listings, which run on Adopets, and they join to the open data because both use the same animal id. They are embedded from the shelter's platform rather than copied, and every dog links back to its real adoption page. —
"No photograph" covers two different situations. Most of it is process: —% of the dogs here under 30 days are not listed yet, because a stray has to sit out a mandatory hold before it can be offered for adoption, and that is why the unlisted median wait is only — days against — for the listed ones.
The rest is harder to explain. The unlisted share never falls below about a sixth at any length of stay, and — dogs have waited more than a hundred days without appearing on the public adoption site at all. Pancho, at the top of this page, is one of them, and so is a German Shepherd who arrived in July of last year and never got a name. A dog can be held back for medical treatment, a bite investigation, a behavioural assessment, or simply be in a foster home, and none of that is published, so this page does not guess. It only notes that the longest waits on the board belong to dogs the public cannot currently see.
Where a dog arrived without a name, the shelter often gave it one later. — of the dogs here are shown under a name somebody chose for them after they arrived.
Live intakes · Live outcomes · Code · Austin Animal Center adoption listings