Most shops decide what to stock by asking what will sell. We decide by asking what we can rule out and still defend the answer in public. This page is that answer, written down, so you can hold us to it.
Applied in this order, because a beautiful claim about a bad material is still a bad material.
01
What is the thing actually made of, all the way down — including the coating, the gasket, the adhesive and the lid liner? We want a full material declaration, not a marketing word like “non-toxic”. Anything that contacts food or skin has to be identifiable by name. If the manufacturer cannot tell us what a component is, that is an answer in itself.
02
A claim only counts if something measured it. We look for test reports against a named standard, from a lab that is not owned by the brand — and we check that the report covers the product being sold, not an earlier version of it. Where we can measure something ourselves, we do, and we publish the method in the Journal.
03
Who made it, where, and will they say so on the record? We ask for the country of manufacture and the factory relationship. A brand that will not name its own supply chain is asking for trust it has not earned, and we are not passing that request on to you.
No amount of certification elsewhere makes up for these. When we reject a product for one of these reasons, we list it in the shop under Did not pass, with the reason attached.
“Fragrance”, “proprietary blend” or an unnamed coating. If one line of the specification is a black box, the whole specification is.
“Chemical-free”, “hospital grade”, “removes 99% of contaminants” with no contaminant named and no method cited.
Food storage or cookware where hot contents touch plastic. This is the single most common reason we turn down otherwise decent kitchen products.
Any surface that becomes waste the first time it is scratched, on a product otherwise built to last a decade.
A logo earned by one product in a range, displayed on the whole range. Or a standard that tests something other than what is being claimed.
We ask every brand the same questions in writing. Silence, or a redirect to the marketing page, is a rejection.
Evra has working access to university engineering laboratory facilities. In practice that means we can do things a normal reviewer cannot: measure how a pan surface degrades over repeated heat cycling, put a water filter through a controlled challenge test rather than trusting the box, and inspect a material under magnification instead of guessing from a product photograph.
It also means we can read a test report properly. A great deal of what passes for product testing online is one person, one kitchen, one anecdote. A great deal of what passes for manufacturer testing is a real measurement of something slightly beside the point.
We are not an accredited testing house, and we do not present our own measurements as certification. When we say a product is independently certified, we mean somebody else certified it and we checked the paperwork. When we say we tested something, we publish what we did, how many samples, and what we could not control, in the Journal. You are entitled to disagree with our method, which is why we show it.
Where the money comes from. Most products we recommend are affiliate links: you buy from the maker, and we may earn a commission at no extra cost to you. Commission never affects whether a product passes, and we publish rejections for brands with affiliate programmes just as readily. The longer version
Reader requests decide most of what we assess next. Join the list and we will send the write-ups as they are published.