How we score dog gear.

How a WagYears recommendation is made

The internet does not need another list of the ten best dog beds written by nobody about nothing. So here is the whole method: what we measure, what we refuse to claim, and where the number on a product card comes from.

Read this first: we research gear, we do not test it

We are a research-led editorial team, not a testing facility. Nobody here has slept five dogs on five mattresses, or run a season of cooling mats through a heat chamber. So we never write "we tested", "hands on", "in our trials" or "we lived with this for a month" - because it would not be true, and an owner deciding where to put a hundred dollars deserves to know which kind of source they are reading.

What we actually do: read the manufacturer's specification and the actual listing, check what the warranty covers, weigh published reviews by people who genuinely do test, and read peer-reviewed and university veterinary sources for the claims underneath a category.

We also read owner feedback, and it is worth being exact about what that means, because "we analysed thousands of reviews" is a claim most sites make and almost none can support. We do not read every review, we do not follow owners over months, and we run no panel and no survey. What we use is (a) the aggregate star rating and the rating count Amazon publishes on the listing, and (b) a read of the reviews Amazon surfaces on that page, looking for themes that recur rather than for a quotable line. When opinion genuinely splits - firmness, durability, sizing - we report the split. Nothing here tells us who those reviewers were: Amazon's aggregate rating is not restricted to verified purchasers, and we do not treat it as though it were.

One consequence, stated plainly: we will not assert a safety or compatibility property we have not sourced. If a manufacturer does not say a bed is chew-resistant, or that a mat is safe for a dog who chews, we say the manufacturer does not say it. The honest ceiling on our claims is "the maker states this."

The WagScore

Every gear pick carries a WagScore from 0 to 100. Read it narrowly, because it is a narrow thing: it combines a public Amazon rating with our own written judgement about whether the product suits a life stage. It is not a safety certification, not a test result, and not a verdict on build quality. No product on this site has been inspected, loaded, crash-tested or slept on by us, and a high WagScore says nothing about whether a crate will hold a determined dog or a harness will hold in a collision. It is deterministic - the same inputs always produce the same number - and it is made of exactly two things:

WagScore = 0.7 x ownerSatisfaction + 0.3 x stageFit

Owner satisfaction (70%). The aggregate star rating Amazon publishes on the listing, weighted by the rating count behind it, so that a 4.6 across twenty thousand raters counts for more than a 4.9 across nine. It is a public retailer aggregate, not a purchase-verified panel and not our own measurement, and its known weaknesses - incentivised reviews, ratings pooled across variants of one listing, no way for us to see who reviewed - come with it. We use it because it is the largest body of real-world experience available for this equipment, not because it is clean.

Stage fit (30%). Authored by us, per category, per stage. It answers one question: does this product actually serve the need the stage has? A deep orthopedic mattress scores high for a senior dog and low for a teething puppy who will destroy it. A crate with a divider panel scores high for a new puppy and is irrelevant to an eight-year-old. This is the part a rating alone cannot tell you, and it is the reason the same product can rank differently on two pages of this site.

Provisional scores are not shown. Where we could not read a product's owner-feedback data, the owner-satisfaction term would be a neutral placeholder rather than measured evidence. We do not publish a score in that state, and where a rating is missing from a comparison table we print that it is missing rather than filling in a plausible number.

Sourcing rules we hold ourselves to

• No untraceable statistics. This subject area is heavily polluted with confident, precise, invented figures. If we cannot open the primary source, the number does not appear. Where a claim comes from a manufacturer's own study rather than an independent one, we say so in the sentence that carries the claim.
• No citation theatre. A source link goes to the page that supports the claim, not to an organisation's front door.
• Manufacturer claims are labelled as manufacturer claims. Including the flattering ones on products we recommend.
• Every product link resolves to a live listing. Dead and mismatched listings get fixed or the product comes down.
• Nothing on this page describes a process we do not run. If a sentence here says we do something, we do it. Where the honest answer is "we read what the listing publishes", it says that instead.

Price

There is no artificial price cap on this site, because some of these categories genuinely are expensive - orthopedic beds for giant breeds, crates, GPS trackers, ramps. What we do instead is tell you what the extra money buys and when it is not worth it, and we keep a value pick on every page for people who need the thing to work rather than to be the best one made.

Getting it wrong

We will. Prices move, listings change, products are discontinued and we misread things. How to tell us about an error is written out on the how-we-research page, along with what we do about it.

We are not your vet

Everything on this site about health, ageing, mobility, weight or behaviour is general information. It is not a diagnosis and it is not a treatment plan. Those come from your veterinarian, who has met your dog.