How our recommendations work

No black box, no sales funnel. Here's the logic, and the promises behind it.

It's rules, not roulette

The assessment is a fixed set of rules, written down in advance, that runs entirely on your device. It weighs the things that genuinely drive care decisions: how many daily activities need help, falls and mobility, memory changes, safety incidents like wandering, how much support is already in place, and what's financially realistic. Your answers land in one of five outcomes, from "aging in place with a light safety net" to memory care.

Because it's deterministic, the same answers always produce the same result. There's no AI judgment call deciding what to show you, and no algorithm optimizing for what earns us more.

The guardrails we've built in

Safety outranks everything. Crisis signals (like a fall with injury for someone living alone), put "call their doctor today" guidance at the top of the result, above anything we could earn from.
Serious memory signals always get a serious answer. A dementia diagnosis never results in a casual "they're fine" answer, and any wandering episode triggers guidance to ask a physician about a memory evaluation, regardless of which result you land in.
We never diagnose. The assessment reads what you've observed; medical judgment stays with your parent's physician, and the results say so.
Money can't touch the logic. Partners never see the rules, can't pay to change them, and the guidance text is identical whether or not an affiliate link exists for it.

Why we don't publish the full rubric

We publish the logic and the promises, but not the exact thresholds, for one reason: gaming a care assessment helps no one, and the thresholds get tuned as we learn from families using it. What never changes is the deal: your answers decide the advice, and nothing else does.

Where the numbers come from

Cost figures come from the CareScout Cost of Care Survey 2025, localized to your state once we've verified that state's figures (until then you'll see national medians, labeled as such). They're medians; real local prices vary, and every cost module says so. Memory care isn't a surveyed category, so we present it as a labeled estimate relative to assisted living, never as a survey figure. Facility data, when the Facility Checker ships, comes directly from Medicare.gov Care Compare with the retrieval date shown.

How we pick the products we show

Some results include specific products (a grab bar, a shower chair, a pill dispenser) matched to the needs you actually reported in the assessment. A product only appears when your answers point to the problem it solves; if your answers point to a level of care where do-it-yourself equipment isn't the honest answer, no products appear at all.

We select on specifications, safety certifications, and documented reliability: weight ratings, how a thing mounts, whether a certification exists for its category, and whether the design avoids the known failure mode (a suction bar that lets go, a riser that shifts, an uncertified bed rail). We have not hands-on tested these products, and we don't pretend otherwise, which is also why you'll never see a star rating from us. Each need gets one pick and at most one alternate. The picks earn us a commission at no cost to you, and the commission never changes which product is matched; the criteria above do.