How we pick

We define the buyer situation, gather products that are currently sold, compare published specifications and materials, read what owners report over months of use, and select on criteria we state on the page.

We do not run a test lab. We have not personally used every product on this site, and we never write as though we have. Where a judgment comes from published specifications or from owner reports rather than our own hands, the page says so.

That statement is the reason this page exists. It is easy to write in the first person about products you have not touched, and a lot of sites in this category do. We would rather tell you what our recommendations actually rest on, so you can judge them rather than take them on faith.

Every roundup, comparison and review here carries a What we could not verify section for the same reason. It is the honest version of the confidence a testing claim implies.

1. Define the reader's situation first. Not 'best shaker' in the abstract, but 'what should someone whose tins stick actually buy'. The criteria come out of the situation, not out of the product catalog.
2. Gather currently-sold candidates. We check that a product is still available at the time of writing. A roundup recommending a discontinued product is worse than no roundup.
3. Compare published specifications and materials. Dimensions, capacities, steel grade, construction method, what is in the box. Numbers we can point at rather than adjectives.
4. Read owner reports at volume. Not to count stars, but to find the failure modes that only show up after months - the finish that wears at the join, the reservoir that needs cleaning, the shelf that sags.
5. Pick, and name the trade-off. Every product on this site carries a 'skip this if' line. It is mandatory, and it sometimes costs us the click.
6. State what we could not verify. Explicitly, on the page, in its own section.

The criteria differ by category because the failure modes differ. These are the axes we use, and they are the ones you will see referenced in each roundup's 'how we picked' section.

We earn a commission when you buy through a link on this site, at no extra cost to you. That is how the site is funded and there is no other revenue - no sponsored placements, no paid inclusions, no gifted products.

Commission rate has never determined an ordering on this site, and no product appears here because of what it pays. Almost everything we link is at Amazon, which means the commission differences between the products on a page are small or nil - the incentive to game an order barely exists.

We also do not accept free products. If that ever changes, this page changes first and any page affected will say so.

The full commercial disclosure is on affiliate disclosure.

You will not find a number next to any product on this site. Prices change constantly, and a figure baked into a page is wrong within weeks - which is worse than no figure, because it looks authoritative.

Instead, every price link opens the current listing at the retailer. Where we need to describe where a product sits, we do it in words - entry tier, mid tier, the step up, the one worth stretching for - which stays true regardless of what happens to the price this week.

We also never copy a price from a competitor's page or a search result and present it as our own. If we cannot verify a number, we do not print it.

If something here is wrong, tell us and we will fix it. Corrections are noted on the page rather than silently edited, and the fastest route is contact.

We revise a page when a product is discontinued, replaced by a successor, changes specification, or is overtaken by a better option - and we recheck every roundup on a regular cycle for exactly the first of those. A roundup recommending something you cannot buy is the most common failure in this genre.

The full correction and update policy is on editorial policy.

The honest answer is first-hand use. A site that has genuinely lived with these products for months can say things this one cannot, and we would rather build toward that than pretend we already have.

When that happens it will be labeled clearly and it will be specific - which units, for how long, and what we actually observed. Until then, every page states what its judgments rest on, and we would rather you hold us to that than trust a claim we have not earned.