Shelters are not short on care. They are short on clean handoffs.
The teams doing this work carry an amount of context, intake facts, foster texts, vet notes, photos, applicant questions, follow-up promises, that was never going to fit cleanly in a spreadsheet or a shared inbox. This page describes what operations research and hard experience say about that problem, and how the record is designed around it.
What the field knows about the work.
These are not slogans. They are the observations that shaped every design decision in the workspace, stated plainly enough to argue with.
Context scatters, and scattering is the failure mode
An intake note in a folder, a foster update in someone's phone, a vet record as an unnamed scan, an applicant's question in a personal inbox, a public listing with week-one wording. Each is true and each is somewhere else. The operational problem is not that teams lack information; it is that the information never meets in one place where a decision is made.
A handoff is only as good as what travels with it
Intake to foster, foster to reviewer, reviewer to public profile, applicant to decision: every handoff in shelter work carries context across people and time. When the context arrives incomplete, the next person decides with a fragment. The record exists so the fragment stops being the unit of work.
Unknown is a valid state, and guessing is the expensive error
Marking 'comfort around young children' as unknown because nobody has seen her with young children costs nothing. Guessing costs a failed placement, a returned animal, and a family that will not adopt again. The system treats unknowns as first-class state, visible and explicit, never as blanks to be filled by inference.
Follow-up is a queue, not a memory
Medical follow-ups, vaccination dues, foster check-ins, review tasks: each has a stored date and a named owner. The Today queue reads those dates and states and shows what is due, overdue, and blocked. It is never scored, ranked, or predicted. Dates and ownership are the whole mechanism.
How one record answers the scattering problem.
One record, every surface
One record, every surface
Intake, care plans, medical and behavior history, foster coordination, applications, public profiles, tasks, communications and reporting all attach to the same animal record. They are surfaces of one thing, not separate tools with separate logins and separate truths. When a foster check-in arrives, it lands on the animal it belongs to, routed to a reviewer, with the animal's full context beside it.
Review gates hold the risky transitions
Review gates hold the risky transitions
Intake carries a review gate a named reviewer must clear. A public profile is drafted, reviewed by a person, and versioned before it goes live. A materially relevant foster check-in re-flags a published profile for re-review, and the profile is held until the review clears. The gate is the mechanism that keeps a fast-moving record honest.
Owners and due dates, everywhere
Owners and due dates, everywhere
Every next step carries the person responsible and when it is expected. Work items assign to team members or volunteers. Holds carry an owner and a review date. The audit trail records who changed what, when, why, and from which source, append-only, so accountability is a property of the system rather than a virtue of the staff.
Foster context that actually reaches the record
Foster context that actually reaches the record
Foster households, placements, and check-ins that set the next expected check-in. Foster notes carry an urgency flag so a question does not wait behind routine updates, and a foster-raised need becomes a placement-planning input rather than a text that scrolls away. The foster board plans placements; the record keeps the thread.
Applicant review, with people deciding.
Applications move through explicit stages with a recorded history. Every stage change requires a named human decision owner, and approvals, conditional approvals and declines require a structured reason. No automated scoring, no AI-driven outcome, enforced in the domain rules, not in a policy document.
What the pipeline guarantees
- Whose move it is, and what is missing: applications carry next steps, missing-document flags, and due dates instead of disappearing into inboxes.
- A complete stage history, appended, never rewritten: every transition is an event with an actor, a time, and a reason.
- Approval-gated communication: no email sends itself. Outbound messages wait for a named person to approve them first.
- The applicant's own view: applicants track their status through a secure portal instead of calling to ask.
WhiskerMatch does not score, rank, or approve anyone. Saving interest is not an application, and an application is not a decision.
What we do not know yet.
Organizing work is not the same as improving outcomes.
We believe cleaner handoffs lead to better placements, shorter stays, and fewer returns. That is a hypothesis, not a finding. No shelter has completed a pilot, so we have no outcome data, no length-of-stay numbers, no return rates. We will not claim improvements we have not measured. The system is built to make those measurements possible: the same records that run the work are the records a study would need.
Follow-up cadence, review staffing, and gate placement are open questions.
How often should a foster check in? Which transitions genuinely need a gate, and which gates just add delay? We have defaults shaped by operator judgment, not by trials. Real pilots will change them, and that is the point of the pilots.
Small teams have different constraints than large ones.
A two-person rescue and a municipal shelter do not have the same failure modes. Much of the system's design assumes a team with named roles; we are still learning where that assumption breaks for the smallest organizations, and what the record should do instead.
If you run shelter operations and the design gets something wrong about how the work actually moves, that correction is worth more to us than a pilot signup. Tell us.
Bring the handoff that keeps breaking.
Intake, foster updates, applicant review, profile review: bring one real workflow, and we will map it against the record in a single conversation.