Live Release Rate Calculation and the 90% No-Kill Threshold

The Formula Behind Every No-Kill Headline

Every live release rate calculation follows the same Asilomar-standardized formula: live outcomes divided by total outcomes, expressed as a percentage. Live outcomes include adoptions, transfers to other agencies, and returns to owner. Total outcomes add euthanasia (all categories) plus death-in-care plus lost-in-care to the live numerator. Get the math right and you have a defensible, donor-trustable headline number. Get it wrong, or game it, and the reputational risk eventually catches up with the organization.

The 90% live release rate threshold is the Best Friends Animal Society “Save Them All” benchmark for no-kill status, and it has become the dominant industry definition since the early 2010s. The Best Friends Network publishes a No-Kill Tracker covering every reporting community in the United States, ranking shelters and identifying where the lifesaving gap remains.

This guide walks through the calculation step by step, explains what 90% does and does not mean, and covers the four ways shelters commonly distort the metric — accidentally or otherwise.

Walking Through the Live Release Rate Calculation

Take a sample year: a shelter records 800 adoptions, 100 transfers to other rescues, and 50 returns to owner. That’s 950 live releases. The same year, 50 euthanasias and 5 in-care deaths produce 55 non-live outcomes. Total outcomes: 1,005. Live release rate calculation: 950 divided by 1,005 equals 0.945, or 94.5 percent. The shelter exceeds the 90 percent no-kill threshold by a comfortable margin.

The numerator includes any outcome where the animal left your care alive: an adopter took them home, another rescue or shelter accepted a transfer, or the original owner reclaimed them. The denominator includes everything in the numerator plus every animal that died in your care — whether by planned euthanasia, in-cage death, or anesthesia loss. Animals still alive in your care at the close of the reporting period are excluded from both numerator and denominator, which is why annual LRR reports stabilize once intake and outflow are seasonally aligned.

What the 90 Percent Threshold Actually Means

Best Friends and the broader no-kill movement set 90 percent as the threshold based on a defensible premise: across a population of shelter animals, roughly 90 percent are healthy or treatable, and 10 percent will face medical conditions of irremediable suffering or severe public-safety dangerousness that ethically support euthanasia. The 90 percent line therefore represents saving every animal who can reasonably be saved.

The threshold does not mean zero euthanasia — and rescues that pretend otherwise mislead donors. Animals with terminal cancer experiencing severe pain, dogs with documented unprovoked attack histories on humans, and neonates born with fatal congenital conditions are ethically and operationally appropriate candidates for euthanasia. These cases are tracked in distinct Asilomar categories (owner-requested euthanasia for terminal-illness surrenders, untreatable euthanasia for medical, dangerous-dog euthanasia for behavioral) and are excluded from the “preventable euthanasia” calculation in some reporting frameworks.

Read LRR Alongside Length of Stay

A high live release rate calculation in isolation can hide problems. A shelter reporting 95 percent LRR while running 90-day average length of stay is moving animals out alive but warehousing them through extended stays that produce kennel stress, disease risk, and behavioral deterioration. The 95 percent looks great until you visit the building and see what 90 days of confinement actually does to a shy cat.

Best practice is to publish LRR, median LOS, and the Asilomar nine-category outcome breakdown together in a single monthly transparency report. That triad is what sophisticated donors, foundations like Maddie’s Fund, and accreditation bodies actually look at. Headline LRR alone is increasingly treated as a marketing number, not an operational one.

The Four Common Ways LRR Gets Gamed

The first and most common distortion is heavy transfer activity to other shelters: every transfer counts as a live release at your organization, but if the receiving shelter euthanizes the animal a month later, the lifesaving was illusory. Transparency-minded rescues using the pull-from-shelter rescue model now report transfer-out and transfer-in volumes alongside LRR and track downstream outcomes where possible.

The second is managed admission so aggressive it shifts the burden onto other organizations or onto owner self-euthanasia. The third is reclassifying preventable euthanasia into the “untreatable” or “owner-requested” buckets without medical or behavioral documentation supporting the call. The fourth is excluding animals never officially intaken — strays held in “stray hold” status off the books, animals seen in the parking lot and never logged. All four distortions show up under serious audit and damage donor trust when exposed.

Reporting LRR Publicly

Best Friends Network’s Save Them All campaign aggregates community-level LRR for every reporting jurisdiction in the United States and publishes interactive maps showing which communities have achieved no-kill status. Submitting data to Best Friends and to Shelter Animals Count puts your organization in the public benchmark conversation and qualifies you for visibility in their grant and resource networks.

Publishing on your own website is equally important. A dedicated transparency page with monthly intake, outcome, and LRR breakdowns — using standardized Asilomar definitions — signals operational maturity to major donors. Most foundation grants now require this level of reporting as a precondition of funding.

When LRR Is the Wrong Metric

For some rescue models, LRR is not the most informative metric. A pure transport rescue moving animals from over-capacity Southern shelters to Northeast adoption markets will show near-100 percent LRR by construction — they only accept animals they’re confident they can place. The meaningful metrics there are throughput volume, partner-shelter LRR improvement, and post-adoption return rate.

A medical-rescue specializing in catastrophic cases (hit-by-car, severe neglect, end-stage chronic disease) may legitimately run lower LRR than a generalist shelter — and the lower number reflects mission, not failure. Context matters. Always read LRR alongside intake population, mission scope, and time-series trend rather than as a single benchmark to be hit at all costs.

Community-Level vs Organization-Level Live Release Rate

The Best Friends “no-kill community” designation applies at the community level, aggregating every reporting shelter and rescue in a defined jurisdiction (usually a county or city). A community can achieve 90 percent LRR even when individual shelters within it run lower numbers, because the broader network of foster-based 501(c)(3) rescues pulls animals from the higher-intake municipal shelter and places them at near-100 percent LRR.

This community-level framing matters operationally. A municipal shelter running 82 percent LRR is not failing if it’s the front door for a community where private rescues pulling from it together produce 91 percent overall. The municipal shelter’s mission is open-admission intake; the private rescues’ mission is selective placement. The two complement each other, and the right metric is the community total. Best Friends Network’s No-Kill Tracker explicitly aggregates at this level.

Tracking LRR Improvement Over Time

The single most powerful LRR data presentation is the multi-year trend chart. A rescue that moved from 78 percent in 2020 to 85 percent in 2022 to 91 percent in 2024 tells a story of operational maturation and lifesaving investment. The annual snapshot — “we’re at 91 percent” — is less compelling than the trajectory. Build the trend chart into your annual report, your major donor presentations, and your board dashboard.

Trend analysis also catches problems early. A rescue running steady 92 percent LRR for three years that drops to 88 percent in the current year has a problem worth investigating: a new disease outbreak in the cat population, a staffing change in adoption operations, a fundraising shortfall constraining medical care for treatable cases. The trend exposes what the single-point number hides.

Frequently Asked Questions

What counts as a live release in the formula?

Live releases include adoptions to permanent homes, transfers to other rescues or shelters (assuming legitimate transfer to qualified partners, not dumping), and returns to original owners after stray intake. Foster placements are not outcomes — the animal is still in your care until adopted or transferred from foster.

Do we count untreatable euthanasia against our live release rate?

Per strict Asilomar formula, all euthanasia counts in the denominator. Some reporting frameworks separately publish “preventable euthanasia rate” that excludes untreatable cases, but the headline LRR per Best Friends and Shelter Animals Count includes all in-shelter deaths. Be explicit in your public reporting about which definition you’re using.

How does the 90% threshold apply to small foster-based rescues?

Foster-based rescues typically run LRR above 95% because they’re selective about intake and rarely euthanize. The more meaningful metric for foster-based organizations is intake volume relative to community need — a 100% LRR rescue that intakes 20 animals a year is contributing less than a 92% LRR shelter intaking 2,000. Pair LRR with volume in all reporting.

Where can I see other shelters’ live release rates?

The Best Friends Animal Society No-Kill Tracker publishes community-level data for the United States. Shelter Animals Count maintains a national database with individual organization breakdowns. Many state shelter associations also publish annual reports. Use these resources to benchmark your shelter against peers in similar markets, and link your organization’s data alongside the no-kill movement context when explaining metrics to donors.