Our methodology
No hedging, no “it depends, buy whichever.” A verdict is earned through research you can trace on the page itself and the products we reject stay published, with reasons.
Two verdicts. Nothing in between.
It earned its place
It does what it claims at its price, holds up according to the evidence we gathered, and we can tell you exactly who it’s right for and who it isn’t. Every KEPT still names at least one real drawback, because no product is perfect and pretending otherwise is how trust dies.
Skip it here’s why
A CUT isn’t a hit piece; it’s a documented reason: an inflated price for what you get, a durability pattern in owner reports, a claim the specs don’t support. Cut products stay on our pages, because knowing what to avoid is half the value of a recommendation.
How a verdict gets made
Define what matters first
Before we look at a single brand, we write down the criteria that actually decide the purchase in that category the sodium content of an electrolyte drink, the return policy on a mattress, the stitch density on a boot. Criteria come first so the field can’t be reverse-engineered around a favorite.
Build the field honestly
We assemble the candidates a real shopper would actually be choosing between: the category leaders, the budget options, and the products whose marketing is loudest. Each roundup states how many products were considered, not just the winners.
Research like an owner
We normalize manufacturer specs so a number on one brand’s box means the same thing as the number on another’s. We read long-term owner reports across multiple retailers the three-month and one-year reviews, not the day-one impressions. We check expert and industry sources where they exist, and we date every price so you know exactly when it was true.
Say plainly what we did and didn’t do
When a page reflects hands-on time with a product, the article says so explicitly and describes what was actually done and for how long. The “How we chose” section of every article tells you its exact basis. We never dress up desk research as field testing.
Publish the verdict, then re-check it
Every evaluative page ends with the verdict, who it’s for, who it’s not for, and what got cut. Verdicts are re-reviewed roughly every six months, and immediately when a product is reformulated, discontinued, or its price moves enough to change the answer. When a verdict changes, we say what changed and why the full process is in our editorial standards.
“If an article doesn’t say we used a product, we didn’t.”
the rule every KeptList page followsWhat gets a product cut
A price that isn’t justified by any measurable difference from cheaper options.
A consistent failure pattern in long-term owner reports durability, leaks, sizing, support that evaporates.
Marketing claims the label, specs, or available evidence don’t back up.
A “premium” version whose only premium is the markup.
Anything we wouldn’t hand to a friend who asked, “just tell me what to buy.”
Built so verdicts can’t be bought
Reader-supported
Some links are affiliate links we may earn a commission if you buy, at no extra cost to you. That’s the business model, in one sentence.
Money never votes
Commissions never decide a verdict, the order of picks, or what we cover. Full details in our affiliate & AI disclosure.
Cuts stay published
The clearest proof is structural: our CUT sections earn nothing and stay up anyway. We don’t accept freebies or payment for coverage.
Who’s accountable
Every page is reviewed and approved by a real, named human editor before it goes live. Research and drafting are AI-assisted; verdicts, verification, and final sign-off are human and if we got something wrong, the editor of record answers for it.
Read our editorial standardsAsk about a verdict