Why a side sleeper's mattress has a different job to do

Lie on your back and your weight spreads across your shoulder blades, your buttocks and your heels. Roll onto your side and almost all of it lands on two points: the shoulder and the hip.

That is not a marketing claim. It has been measured with pressure mats, and the difference is larger than most people expect.

What actually changes when you roll over

Pressure mapping across three sleeping postures found that moving from your back to your side reduces contact area by about 18% while raising average pressure by roughly 50%.¹

Same body, same mattress, same night. Turn onto your side and the same weight is carried on a smaller area, concentrated over two bony points rather than spread along the length of you.

What that asks of the mattress

On your side

Your shoulder and hip stick out further than your waist. If the surface is too firm, those two points cannot sink, so your spine bends sideways in the middle. If it is too soft, your hip sinks too far and the spine bends the other way.

This has been measured. A study of 25 men tracking markers on twelve vertebrae found lateral spinal deviation of 12.66° on a soft surface, 8.9° on a firm one, and 4.10° on a surface zoned to the individual.² The difference between the extremes is roughly three times.

On your back

The problem inverts. Your body is broadly level, but there is a hollow under the lower back that the mattress has to fill. Too firm and the lumbar curve flattens; too soft and the whole middle sags.

Measured on 16 people with a 30-sensor curvature tape: compared with a medium mattress, a soft one increased cervical curvature by about 27mm, while a firm one reduced lumbar curvature by about 11mm.³

Where the evidence stops

Here is the part the industry does not mention.

We could not find a trial comparing mattress firmness by sleeping position against any outcome — not pain, not sleep quality, not alignment. That is what our search turned up; if such a trial exists we would like to be pointed at it. The landmark firmness trial — Kovacs and colleagues in The Lancet, 313 people over 90 days — does not include sleeping position as a variable anywhere in the study.⁴ A 2015 systematic review of 24 studies concluded that a self-identified medium-firm mattress was optimal, with no stratification by position at all.⁵

The closest anyone has come is a Belgian sleep-laboratory study of 17 people, recorded by full polysomnography. It deliberately gave people a badly zoned surface — stiff under the shoulder, soft under the waist — and measured what happened. Among the participants who genuinely slept mostly on their side, wake time rose and REM sleep fell significantly. Among those who split their night between side and back, nothing changed at all.⁶

That is real evidence that position interacts with the surface. But note what it shows: a deliberately bad shoulder zone harms committed side sleepers. It does not show that mattresses marketed at side sleepers outperform anything.

The firmness numbers are not comparable

Every retailer publishes a scale. Side sleepers 4 to 6 out of 10. Back sleepers 5 to 7. Firmer if you are heavier.

We could not find a standard behind any of it. BS EN 1957 is the only relevant British standard, and it measures hardness by pressing a rigid pad into the mattress. Its published scope says, in terms, that “it needs to be emphasized that the firmness rating cannot be used to demonstrate comfort and/or quality of a mattress or a complete bed”, and it makes no reference to sleeping position whatsoever.⁷

Two things follow that are worth knowing before you compare anything.

The standard's scale runs backwards from the shop's. In the Kovacs trial, firmness was measured on a scale where 1 is firmest and 10 is softest — his "firm" mattress scored 2.3 and his "medium-firm" 5.6. Retail scales run the other way. Anyone trying to reconcile the research with a shop's chart will read it inside out.

UK retailers do not agree with each other. Bensons for Beds publishes 6 for side sleepers and 6.5 to 7 for back sleepers.⁸ Others publish 4 to 6 for side sleepers. We could not find a cited source for either. Since no standard defines the scale, both may be perfectly reasonable in-house judgements — but they are not measuring against a common reference, and a reader comparing them is not comparing like with like.

The same applies to ILD, the foam firmness figure sometimes quoted on spec sheets. The international standard defines five different test methods, and states plainly that results "relate only to the test conditions specified and cannot, in general, be used directly for design purposes."⁹ Two brands quoting ILD may not have measured the same thing, and a single foam layer's rating tells you little about a finished mattress with a spring unit under it.

Broad shoulders do not need a softer mattress

This one is worth singling out, because it is repeated everywhere and a British study found the opposite.

Researchers at the University of Central Lancashire measured 59 people in side-lying with infrared motion capture, testing which body measurements predicted the firmness that kept the spine straightest. Body weight, height and hip circumference all predicted it. Shoulder width showed no differentiating effect. Neither did BMI.¹⁰

Larger hip circumference was the finding that mattered: it produced significantly greater spinal deviation on softer surfaces. A separate study of 150 people across a range of ages and body types concluded that medium firmness gave the most consistent spinal support across the lot of them.¹¹

And you do not sleep in one position anyway

The whole premise assumes you pick a position and stay there. Accelerometer data says otherwise.

A study of 664 adults wearing sensors in their own beds recorded 1.6 position changes per hour — around a dozen across a night.¹² Video observation of 53 people found 18 to 24 posture changes per night.¹³

Even committed side sleepers are not committed. In the Belgian study, the most side-dominant group still spent 61% of the night on their side — and 38% in other positions. The other group split roughly half and half.⁶

Your mattress has to work in every position you use, on the same night. A surface optimised hard for one of them is being optimised for perhaps 55 to 60% of your sleep.

What the rules say about all this

Worth knowing what a shop is allowed to tell you. The Advertising Standards Authority upheld a complaint against Origin Sleep in December 2024 over claims including "orthopaedic back support", "engineered for pain-free sleep" and "the highest possible comfort and spinal alignment". The ASA's reasoning was that it expected "robust documentary evidence... consisting of clinical trials conducted on people", and none was produced.¹⁴

Since no position-stratified clinical trial exists, claims that a particular mattress corrects a side sleeper's spinal alignment or relieves shoulder pain cannot currently meet that bar. "Orthopaedic", incidentally, is not a protected term and carries no certification.

What to actually do

• Know your dominant position, but do not obsess over it. It is one input among several, and not the strongest.
• Weight and hip measurement matter more. Heavier or broader through the hips generally means firmer; lighter and shorter generally means softer.
• Ignore the number out of ten. It is not defined by anything, and two shops will give you different ones for the same body.
• Use the trial properly. This is the only real test available to you, and it is free. Sleep on it for the minimum period the trial requires, in your own bed, in all the positions you actually use.
• Check your spine feels level on your side. Have someone look at you from behind. If your waist dips or your hip rides high, the surface is wrong for you — regardless of what it says on the label.

Persistent back, neck or shoulder pain is a matter for a GP or physiotherapist rather than a purchase. No mattress sold in the UK is licensed to treat anything.