You Can’t Fix What You Can’t See

California Stopped Counting Its Shelter Animals, and Let Private Shelter Data Dashboards Fill in the Gaps

Earlier this month, Animal Politics asked what happens when the field’s “neutral” scorekeeper, Shelter Animals Count (SAC), becomes a program of one of the industry’s most powerful and controversial charities, the ASPCA. That piece focused on governance, modeling, and a 147,000‑animal gap in SAC’s own national tables.

Those aren’t abstract fights over spreadsheets; SAC’s national dashboards and Best Friends Animal Society’s new Shelter Pet Data Alliance dashboards have become the scoreboards lawmakers, funders, and national groups use to judge performance and shape policy across the field.

Together, the ASPCA and Best Friends now oversee the private systems that, in practice, stand in for missing or fragmented state-level shelter data.

Earlier critiques examined SAC’s national modeling and governance. This time, the lens narrows to one state, California, not because it is unique, but because it shows what those choices mean on the ground in a state that stopped collecting its own standardized shelter data and now relies on private dashboards to tell its story.

For years, SAC’s reports have been treated as if they describe “what’s happening in U.S. shelters.” In California, that assumption breaks down. What we have instead is a patchwork: a voluntary national database with partial participation, a state that abandoned its own shelter data collection a decade ago, and local agencies that publish what they choose, when they choose, if they publish at all.

California’s shelter data: partial, voluntary, and largely unseen.

How much of California does SAC actually see?

According to SAC’s own tools, California has roughly 293 shelters in its universe (government‑run agencies, facilities with government contracts, and independent shelters without contracts). That universe reflects the organizations SAC has identified and categorized; it is not necessarily a complete list of every public or private shelter operating in the state. Of those, only 110 submitted any data for 2024. Just 68 of those 110 are government agencies or shelters holding government contracts.

Advocates who have sampled SAC’s California participation estimate that for government and government‑contract shelters alone, only about one‑third reported full‑year 2024 data, leaving roughly 70–75% of intake and outcome activity from those agencies unaccounted for in the SAC database.

In practical terms, California’s “state number” in SAC is built from less than a quarter of the shelters in its own identified universe, and almost certainly from far less than a quarter of actual statewide intake.

Put differently:

• About 37–38 percent of California shelters in SAC’s universe reported any 2024 data at all.​

• Only about 23 percent of those shelters were government or government‑contracted.​

• Roughly 62–63 percent of shelters in the SAC universe did not report 2024 intake/outcome data, and many appear to have never reported in any year.

A 2023 participation snapshot, combining SAC’s “participating organizations” view with a hand‑built list of California agencies, underscores the same point. Statewide, only a modest fraction of municipal agencies and contractors have ever submitted full‑year data, and some large jurisdictions, including high‑euthanasia operations, either never appear or drop out after a few years.​

When you see a California number in a national SAC report, you are not seeing “California.” You are seeing whatever slice of agencies chose to report that year, with no way for an outsider to adjust for who’s missing.

That means a county with rising intake and euthanasia rates can disappear from the state’s “progress” narrative if it does not report, while better-resourced or lower-intake shelters stand in for California as a whole. Any statewide analysis built on that slice risks overstating progress, understating crisis, and misdirecting resources away from the counties that need them most.

From dashboards to black box

Coverage isn’t the only problem. Access and functionality have changed.

Until recently, SAC offered dashboards that allowed users to:

• Filter by state, shelter type, and time period.

• Examine euthanasia trends over multiple years for a given state.

• View state‑level intake and outcome breakdowns with some ability to interrogate what was driving change.​

Those views made it possible, with effort, to reconstruct trends for a state like California and at least understand what the reporting universe looked like. Advocates could compare state‑level numbers to local reporting from large shelters and see where the gaps were.

In 2024, SAC consolidated much of this into a National Animal Welfare Statistics Dashboard, a slick, aggregated view that no longer allows users to filter by state, county, or shelter type. California still appears, but only as a modeled slice of a national picture, without the tools needed to understand which agencies are driving the numbers, or which are missing entirely.​

A separate “state‑level” dashboard exists, but it has lost key filters: you can’t break out data by animal type, or by shelter vs. rescue, in the way you could before. On the state‑level dashboard, only high‑level aggregates now appear; the detailed filters by agency, animal type, and intake source that were available in 2023 are gone, and the dashboards do not distinguish between true reductions in intake and animals turned away, diverted, or abandoned due to limited‑intake policies.

For anyone trying to understand, for example, whether an apparent euthanasia increase is driven by large‑breed dogs in government shelters or neonatal kittens in rescue networks, that’s a serious loss of analytic power.

Gatekeeping the Data

At the same time, access to the underlying, organization‑level datasets has shifted behind a “Request the Data” gate: applicants must explain how they intend to use the data, some requests involve fees, and SAC has periodically closed the request pipeline altogether with notices that requests are “closed for the remainder of the year.”

A sample state‑level file remains available as a free download, but broader access to granular data now depends on SAC’s approval rather than simple public download. For data that largely originates in publicly funded shelters and publicly reportable activity, that represents a meaningful change in who can independently test or challenge the story the dashboards are telling, especially now that SAC is a program of the ASPCA rather than an independent data trust.

The timing is notable. As SAC touts its ASPCA acquisition as a guarantee of “clean and independent” data, the public‑facing tools that once allowed users to interrogate that data at the state level have become more opaque and less granular.

When access to granular data narrows at the same moment governance consolidates under one of the field’s largest funders and policy actors, questions about independence are no longer abstract; they become structural.

The result is a double opacity:​

• The coverage is limited and voluntary.

• The tooling that once allowed serious users to see around those limits has been dialed back.

Animal Politics has asked SAC to explain the rationale, criteria, and fee structure for these data requests and will report any response in a future update.

The state stepped away, and never came back

Part of the problem is structural and homegrown in California.

From the mid‑1990s through roughly 2016, the California Department of Public Health (CDPH) collected limited, voluntarily submitted data from public animal shelters. The focus was rabies control: licensed dogs and cats, vaccinations, and bite reports, not a comprehensive picture of intake, outcomes, or euthanasia.​

Around 2016, that state‑level collection effectively ceased. No mandatory, uniform system replaced it.

Since then:​

• National nonprofits have offered voluntary data platforms, most prominently SAC and now Best Friends’ Shelter Pet Data Alliance, but participation is uneven and largely unverified.

• There is still no complete, centralized list of California’s publicly funded animal shelters; even advocates working on spay‑neuter legislation and transparency bills have had to assemble their own spreadsheets from CalAnimals membership lists, local government directories, and on‑the‑ground knowledge.​

• Legislative efforts to restore or modernize state‑level reporting, such as bills to resume shelter data collection as part of rabies reporting, or to require monthly publication of intake and outcome data, have repeatedly stalled or died in committee.

California, in effect, outsourced much of its visibility on shelter operations to private, voluntary systems and then treated those outputs as if they were official.

And SAC is not alone: Best Friends Animal Society’s Shelter Pet Data Alliance has emerged as another de facto national data source, built from a mix of reported and modeled numbers, with limited public access to underlying shelter‑level data. Together, these tools now shape the public story about “progress” while leaving most of the raw material about what individual shelters actually did, and which shelters never reported at all, largely out of public reach.

The animals who never make it into the numbers

Even if every municipal shelter in California reported to SAC, the picture would still be incomplete. A great deal of animal movement never passes through a government facility at all.

Front‑line rescues describe a world of “community intakes” that sit entirely outside shelter statistics:

• Larger dog rescues that take hundreds of dogs a year directly from the streets or from people who never contact a shelter.

• Cat rescues that handle thousands of animals annually through colony management, trap‑neuter‑return, and private surrender diversion, again, without a public shelter ever opening a file.

Some of this activity appears in IRS filings as aggregate annual intake numbers for 501(c)(3)s, and there are commercial services that aggregate those filings into nonprofit datasets. But there is no system that ties those numbers to municipal intake, no statewide registry that distinguishes between small foster‑based groups and high‑volume rescue facilities, and no mechanism for reconciling what public agencies report with what the rescue sector is actually doing.​

Without integrating municipal and rescue data into a single reporting framework, California cannot know whether intake is truly declining, whether animals are being diverted before they reach shelters, or whether they are simply disappearing into informal, untracked channels altogether.

The gap matters. Even if the majority of animal intakes historically occur within shelters, the absence of a complete, verified list of those shelters and a mandatory reporting requirement means that any national dataset will necessarily miss a substantial portion of shelter intake, as well as an unknown but significant volume of community-level rescue diversion.

Animal Rescuers for Change, a California‑based grassroots coalition representing roughly 95 nonprofit rescue and animal welfare organizations statewide, formed in response to the current overpopulation crisis and has begun documenting these “invisible” intakes and community saves, but in the absence of a state‑run framework their work remains fragmented and largely disconnected from official shelter statistics.

Their approach is explicitly pro‑shelter and oriented toward collaboration rather than blame, with the goal of bringing frontline rescue experience into legislative discussions about systemic fixes.

In future articles, Animal Politics will look more closely at how Animal Rescuers for Change is documenting these gaps and pushing for statewide standards in California.

Why this matters for policy, and accountability

None of this is an argument against SAC’s efforts. A voluntary, national dataset is better than nothing. The problem arises when incomplete, partially modeled numbers are treated as authoritative, especially in a state that has chosen not to build its own reporting infrastructure.

For California, the consequences are concrete:

• Legislators trying to understand why euthanasia is rising in some counties are looking at modeled, partial numbers that may exclude exactly the jurisdictions they are hearing the most complaints about.

• Public‑health officials trying to assess the risk of parvo, distemper, or zoonotic disease spread are working without standardized, statewide shelter intake and outcome data, despite clear evidence that unmanaged overpopulation and managed‑intake policies are contributing to disease clusters and mass euthanasia events.

• Communities asking basic questions, such as: “How many animals is my county shelter killing?” “Are we doing better or worse than three years ago?”, often cannot get a straight answer from local websites, and the national dashboards no longer provide a way to drill down.

Meanwhile, rescue organizations report increasing administrative barriers to pulling animals from some of the very shelters that are least transparent about their outcomes: courtesy pull limits, partner‑only euthanasia notifications, rescue‑to‑rescue transfer bans, and rigid enforcement of paperwork requirements that go beyond anything in statute. Those accounts deserve, and will get, their own detailed treatment in a future piece, with documentation and responses from the agencies involved.

In the earlier article on the ASPCA’s acquisition of SAC, Animal Politics argued that the deal raised hard questions about who controls the national narrative. California’s missing numbers raise a related one: what happens when those narratives are built on a foundation that doesn’t actually include most of the state’s shelters?

A state that does not require its shelters to count cannot claim to manage its animal welfare system. And a national dashboard built on partial participation cannot substitute for public accountability.

Until lawmakers require standardized intake and outcome reporting from every publicly funded shelter, and recognize community-based rescue as a structural part of the system, not a rounding error, no national dashboard, however sophisticated, will show policymakers what they most need to see: how many animals enter the system, where they go, and why they die.

Policymakers cannot fix what they cannot fully see.

Ed Boks is the former executive director of animal care and control agencies in New York City, Los Angeles, and Maricopa County, and a past board member of the National Animal Control Association. His work has appeared in the Los Angeles Times, New York Times, Newsweek, Real Clear Policy, Sentient Media, and now on Animal Politics, a lively community spanning 48 states and 61 countries.

If you enjoyed this article, consider subscribing to Animal Politics. Your support fuels independent research, reporting, and discussion shaping the future of animal welfare.