How we pick

Most whiskey gift guides assert that they chose carefully and leave it there. When we fetched nine competitor pages while researching this site, not one of them published how it selected its products. This is ours, in enough detail that you could check our work.

1. Where candidates come from

The Amazon catalog for each category, filtered to items that are actually in stock and shipping. We read live search results rather than working from memory or from other people’s roundups, which is how we avoid recommending a product that has quietly gone out of production.

Every ASIN, product name and image on this site was read off a live listing on 2026-09-02. None of them is remembered and none is invented.

2. What we compile

For each candidate we record what the maker publishes, and only what the maker publishes:

• Capacity - the number that decides whether a decanter takes a full 750 ml bottle, or whether a glass has room for a large ice cube.
• Material - and specifically whether a lead-free or non-lead specification is stated rather than implied.
• What is in the box - glass count, accessories, case.
• Closure - for anything intended to hold a spirit.
• Care instructions - dishwasher safe, hand wash, or unstated.

These come from the listing’s specification section, not from its marketing copy. Where a maker publishes nothing for a field, our tables print not published. We never estimate one, and we never carry a figure across from a similar product. The gaps are visible on purpose: how thoroughly a maker documents its own product is itself a signal, and in this category the variation is enormous.

3. What disqualifies a pick

• Leaded crystal for anything sold as storage. Graziano and Blum measured port stored in crystal decanters going from 89 to 3,518 micrograms of lead per liter over four months. A decanter that will hold a bottle for months needs a published lead-free specification. This is a hard rule, not a preference - the detail is here.
• Missing or evasive material disclosure on a product whose material is the whole question - a “crystal” decanter that will not say what kind.
• No closure where the product claims storage. An open vessel sold as a spirits decanter is a carafe with an optimistic title.
• Chronic stock problems. A recommendation you cannot act on is not a recommendation.
• Performance claims sourced only from the seller. We will still list the product; we will not repeat the claim as though it were independent.

4. How we rank

In this order, and the order is the point:

1. Fitness for the stated giftee and budget. A page headed gifts for beginners ranks on what suits a beginner, which is a different question from what is best in the abstract.
2. Published specifications. Between two similar products, the one that documents itself wins - because you can tell in advance whether it suits your person.
3. Price, last. Cheaper breaks a tie; it never wins one on its own.

Commission rate is never an input. It does not affect which products appear, what order they appear in, or how enthusiastically they are described.

5. What we did not do

We did not test, taste, chill, time or measure anything. There is no lab behind this site, no test bench, and no product in our hands. Not one of the 51 items on this site has been used by us.

We are stating that in the first screen of the page rather than in a footnote because it is the single most important thing to know about our recommendations. Almost every competing guide implies testing it has not done, and the implication is doing a lot of work in those pages.

Where a question genuinely requires measurement - does a stone chill like ice, does lead migrate out of crystal - we cite someone who did the measuring and name them. That is a substitute for testing and we label it as one. The whiskey stones verdict is the worked example.

6. How prices work

Prices are never typed into the copy. They render from a live data layer, stamped with the date they were checked, and they refresh on a schedule rather than at publication.

When that layer has nothing fresh for a product, the button reads Check price on Amazon. That is a first-class state, not a failure state - there is no third option in which we show you a remembered number. Every competitor page we looked at displays prices that somebody typed once and nobody has verified since; that decay is exactly what this design prevents.

The budget pages - under $25, under $50, under $100 - filter their own picks against their ceiling when the page renders. Where the live feed is reporting, they use today’s price. Where it is not, they fall back to the price band each product sat in when we read its listing on 2026-09-02, and the page says so on the page. A band is not a price and we never present it as one.

7. Every pick carries a catch

Every product on this site has a stated drawback, including the top pick on every page. This is enforced by the way the site is built - a product cannot be added without one.

The reason is simple: a roundup where nothing is ever wrong for anyone is an advertisement. If the only signal on a page is how enthusiastic we sound, there is no signal.

8. How to tell us we are wrong

Write to us via the contact page. If a specification is wrong, a product has changed, or a source no longer says what we say it says, we will fix the page and add a dated note at the foot of it explaining what changed. We do not silently edit a claim we got wrong - the corrections policy is here.