Nearly half of all “best mattress” recommendations go to one company

The direct-to-consumer brands that dominate these lists run referral programmes. Nectar’s own affiliate page describes its payouts as among the highest in the industry. This arrangement is not new: in 2017 NPR reported on mattress reviewers being paid per mattress sold, with some earning more than a million dollars a year, and on lawsuits between a mattress company and review sites.

3Z Brands’ six brands account for 48% of every recommendation recorded. Resident Home’s brands account for a further 10%. Somnigroup, which makes Sealy, Tempur-Pedic and Stearns & Foster and owns the Mattress Firm chain, accounts for 2%. Shared ownership does not mean the mattresses are the same, but a list of six “different” brands can be one company six times.

The 3Z Brands figure includes Titan and Plank, which are sold by Brooklyn Bedding. An earlier version of this page counted those two separately, which understated 3Z Brands’ share.

Private equity ownership does not explain the pattern on its own. Somnigroup is the largest company here and is publicly traded; its brands take 2 of the 111 recommendations. Casper was owned by a private equity firm until it was sold to the foam maker Carpenter in 2024, and no guide recommends it. The clearer divide is between brands sold mainly online and brands sold mainly in shops.

One pattern Reccas could not explain: three brands, Amerisleep, Zoma and Vaya, are recommended by only two of the 10 guides, and they are the same two guides each time, Sleep Junkie and eachnight. Sleep Junkie’s own disclosure says the site carries Amerisleep advertising. Reccas has not established who owns Zoma or Vaya, or any connection to eachnight, and draws no further conclusion.

Reccas cannot tell you these mattresses are better; it has not tested any. What it can say is that a guide funded by referral fees has little reason to cover a mattress that earns it nothing, and a mattress bought in a showroom or a warehouse club usually falls into that group. Reccas could not confirm that Costco pays referral fees on products; the descriptions it found mention fees for new memberships only.

Where the tests support the guides. The two mattresses every guide recommends are among the highest scores NapLab has given. Mattresses that at least one guide recommends average 8.80; the rest average 8.48. The brands the guides leave out also score lower:

Where they do not. A high score is not what gets a mattress recommended. NapLab gives 9.0 or more to 69 mattresses that it lists with a US price. The 33 of those made by 3Z Brands collect 37 recommendations from the other guides. The 23 made by smaller independent brands, such as Big Fig, Tiami, Nest Bedding, GhostBed, collect 1.

Resident Home shows the reverse. Its brands collect 11 recommendations, but only 1 of its 23 tested mattresses reaches 9.0. And when a guide does pick from Somnigroup, it picks a low scorer: Sealy Posturepedic Hybrid (7.22) and Tempur-Pedic ProAdapt (7.87).

This does not show that 3Z Brands makes better mattresses. NapLab earns a commission, has tested more mattresses from 3Z Brands than from any other owner, and its score is for a new mattress: it rewards cooling, bounce and trial terms, and does not include wear. NapLab’s separate ten-year wear simulation covers only 17 mattresses, too few to compare owners. Reccas matched 105 of the 109 mattress recommendations to a NapLab review; a few matches are to the closest model.

The brands the guides favour mostly say what is inside. The showroom brands do not, at any price, and neither do Nectar and DreamCloud. A blank here means NapLab’s review lists no figure from the maker; the maker may publish it somewhere Reccas has not looked. Stated density is a fact about construction, not a measure of how long a mattress lasts: in the small set of mattresses NapLab has wear-tested, denser foam did not reliably hold up better.