We asked an assistant, not a panel. Bath and unwind, measured 2026-09-06.

Twenty-five brands answered one question about a hot bath

We asked a grounded assistant what to put in a hot bath to unwind, in two phrasings, and wrote down every brand it named. Here is what came back, what a buyer should do about it, and why we make nothing here yet.

The short answer

Across two phrasings of the question what to put in a hot bath to unwind, no brand was named twice. There is no consensus answer here, so the brand on the label carries almost no information and the attributes do.

This reading is thinner than the others on this site. Only two phrasings returned data for this question set, so "named once" carries less weight here: there was only one chance to be named twice.


Which brands do AI assistants name for what to put in a hot bath to unwind?

Twenty-five distinct brands were named across two phrasings, captured 2026-09-06 against gemini-2.5-flash (google_search grounded). Twenty-five of them appeared exactly once. None appeared more than once, so no brand leads this question. Brands we hold enough history on are linked to their record.

BrandMentionsAlso named for
Asch Building1named here only
Bayou Burns1named here only
Beauty By Earth1named here only
Body Restore1named here only
Cleverfy1named here only
ElizabethW1named here only
Esker1named here only
Fleurco1named here only
Go Easy1named here only
Heatsistence1named here only
JoJowell1named here only
Kallista1named here only

A mention is not an endorsement and nobody paid to be on this list. It is what one grounded engine said on one date.


How should you choose a hot bath?

Because no brand repeats across phrasings, the name on the label is close to noise here. Choose on the attribute you actually need. Any list claiming a single winner in this category is inventing one, because the engines themselves do not agree from one phrasing to the next.

Contestability here is 96 out of 100. That describes how far the field is from having a leader. It says nothing about how many people are asking, and it is not a forecast of anything.

What does a hot bath actually do?

Almost every bath product in this category is sold on one claim: that magnesium from Epsom salt passes through your skin and does something useful once it is inside you. The evidence for that specific claim is thin, and the studies usually cited for it are small, uncontrolled or funded by people selling salt. The odd part is that the category does not need the claim, because the thing a hot bath actually does is well documented and has nothing to do with magnesium.

It moves heat. Immersion raises your peripheral skin temperature and pushes blood toward the surface, and when you get out that heat sheds quickly, so your core temperature falls faster than it otherwise would. A falling core temperature is one of the ordinary physiological signals associated with sleep onset, which is why a bath an hour or two before bed is a commonly recommended routine and why one immediately before bed is less useful: you want to be on the downslope, not still hot.

That mechanism is about water and timing. It does not require anything to be dissolved in the bath at all, which is worth knowing before you spend money on what is dissolved in it.


Does magnesium from Epsom salt absorb through your skin?

The honest answer is that it is not established. Epsom salt is magnesium sulfate, and the marketing claim is that meaningful quantities cross the skin barrier during a soak. The skin is a competent barrier and the studies most often cited in support are small, frequently uncontrolled, and often connected to the industry that benefits.

We are not saying it does nothing. We are saying nobody has shown it does the thing the label implies, and a category where every seller asserts an unestablished mechanism is a category where you should buy on the things you can verify instead.


Salts, bath bombs or oils: what is the actual difference?

Salts dissolve and change how the water feels, slightly slippery at a real concentration and not at all at a token one. They are the format where quantity actually matters, because the effect is a function of how much is in the tub.

Bath bombs are mostly a display product: bicarbonate and citric acid fizzing, plus colour and fragrance. That is fine if the fizz is what you are buying, but you are paying for a performance rather than for the water.

Oils sit on the surface and coat you as you get out, which some people want and which also leaves a film on the tub and can make it genuinely slippery. Worth knowing before rather than after.


Does fragrance matter more in a bath than in a candle?

Yes, and this is the one place in this category where caution is actually warranted. A candle's fragrance reaches you through the air. A bath product's is dissolved in water that is in contact with all of your skin for twenty minutes, at temperature, on skin whose barrier is softened by the soak.

Undiluted essential oils in particular do not disperse in water, so they float and can concentrate against skin. Look for products that state a dispersant or a carrier, and treat a very heavily fragranced bath product with more scepticism than you would the same fragrance load in a candle.


What should you actually check on the label?

Quantity per bath, first. A product recommending a quarter of a cup is selling you fragrance in a salt carrier; a real soak concentration is usually one to two cups, and a bag that only holds three baths at the stated dose is more expensive than it looks.

Then the ingredient order. If magnesium sulfate is not first, you are mostly buying something else. And if the list is magnesium sulfate, fragrance and colourant, understand that the colourant is doing nothing but colouring the water and may mark a light-coloured tub.


What to check before you buy one

Five things that are visible on a label or a listing, in the order they matter.

  1. Magnesium sulfate or sea salt listed FIRST, not third behind fragrance and colour
  2. A stated dose per bath, and a cost per bath you have actually worked out from it
  3. One to two cups per soak, not a quarter cup
  4. A dispersant or carrier stated if essential oils are included
  5. No therapeutic claim on the label: a seller confident in the product does not need one

This list is what we would tell a friend who asked, and it applies to every brand named on this page as much as it does to ours. We publish it because a reader who can judge the category does not need to trust us.


Common questions

Is Epsom salt worth it, or is it just expensive salt?

It changes how the water feels at a real concentration, and it is cheap in bulk. What is not established is the absorption claim used to justify premium pricing. Buying plain magnesium sulfate by the kilogram and adding your own scent is the cheapest version of the same bath by a wide margin.

When should you take a bath if you want to sleep better?

Commonly suggested is roughly one to two hours before bed, so that the post-bath fall in core temperature coincides with when you are trying to sleep. Immediately before bed leaves you still warm, which is the opposite of the signal.

Why do bath bombs cost so much for what they are?

You are paying for the fizz, the colour and the moulding rather than for volume of active ingredient. Bicarbonate and citric acid are inexpensive; the price reflects presentation and packaging. That is a fair trade if the display is what you want from it.


A claim is worth what its instrument is worth. So here is the instrument.

How this page was measured

Captured 2026-09-06 against gemini-2.5-flash (google_search grounded), two phrasings of the same buying question, brand names recorded verbatim. Mentions, not rankings.

phrasings asked
2

The same buying question asked several ways: category, situation, and open.

distinct brands
25

Retailers excluded. Amazon is where an answer gets bought, not a brand competing to be it.

top brand share
4%

Nobody leads. Every brand holds this same share, which is what a field with no incumbent looks like.

contestability
96

How far the field is from having a leader. Not demand, and not a forecast.

There is no ranking to draw here. All twenty-five brands were named the same number of times, so any order would be alphabetical rather than meaningful.

Directional, not decision-grade. Prompt sampling reads an output distribution, and published work on ordered brand lists finds very low run-to-run consistency. We publish the limit because the limit is what makes the rest worth reading.

The shape of the answer

Twenty-five of twenty-five brands were named exactly once.

Every brand here sits on one mention, so there is no split to draw.

When almost every brand appears once, the engine is not reporting a consensus. It is reaching for whatever the retrieval layer surfaced that minute.