About

How we source

This is a hobby site, and it would have been much easier to write it from general knowledge. That version would also have been worthless, because the only thing a THC drink reference is actually for is getting the numbers right.

So here are the rules we built it under.

Every spec was read off a page

Each product in the database links the exact URL its numbers came from and the date we read it. Brand pages outrank retailer listings. Listicles and roundups are not acceptable sources for a spec, no matter how confident they sound, because they are mostly copying each other.

Where only a retailer listing was available, the product page says so.

A gap stays a gap

When a brand does not publish calories, the calorie field says not listed. We do not fill it from a similar product, we do not carry it over from an older formulation, and we do not round it to something tidier.

This is why a fair number of rows are patchy. Of the 101 drinks here, 58 publish a calorie count anywhere we could find it, and 43 do not. Showing you that is more useful than hiding it behind a plausible guess.

A product with no verifiable specification at all is not listed, and several were dropped on exactly that basis. So were two rows that were not really products: a line-wide spec sheet with no confirmed flavour, and a drink still on pre-order rather than actually on sale.

The dose is per serving

The headline milligram figure is always one serving, never the container, and any container holding more than one is flagged as such. Where a brand publishes a per-serving dose but no serving count, the container total stays blank rather than being reconstructed by multiplying two approximations.

There is more on why that distinction matters in Reading the milligram.

Onset claims are quoted, not repeated

Where a brand states how fast its drink works, we record it verbatim and attribute it to them. We do not restate it as fact. The evidence behind the category's speed claims is genuinely thin, which we went through in Do drinks really hit faster?.

The science pages cite public health and peer review

Guides lean on the CDC, state health departments, the American College of Medical Toxicology and peer-reviewed literature. Where sources disagree, the page says so rather than picking the tidier answer. Where the only support for a claim is industry marketing, the page says that too, and there were several.

Some widely repeated figures are deliberately absent because we could not verify them. A specific drug-test detection window for single use is one. A "20 percent lower CB1 availability" figure is another.

What we do not cover

Legality. Hemp THC beverage law varies by state and moves quickly. A stale legal claim baked into 101 product rows would be the most consequential thing on this site to get wrong, so there is no legality field and no map. Check your own state's current rules against a primary source.

Sponsored placement. There is none, and there is no mechanism for it. No brand paid to be included, no brand can pay to be included, and no brand has seen a page before it went up.

We do earn a commission on some links to barware, glassware and mixers, the things the recipes call for. Two things follow that we want to be explicit about. It never affects what is listed here or what we say about it, and it is structurally impossible for it to apply to the drinks themselves, because the retailers that run those programmes prohibit THC products outright. Nothing on this site sells you a THC drink.

When we are wrong

Formulations change without announcement. If the label in your hand disagrees with a page here, the label is right and we are out of date. Every product page carries the date we last checked so you can see how stale a figure might be.