How to choose a mattress: start medium-firm, then fit it to your body

Most mattress advice starts with brands. This guide starts with your body. Here is how to choose a mattress in the order that matters: firmness first, then your body weight and sleep position, then materials, temperature, your partner and the base. Then you test it at home, because no tester, chart or showroom nap can tell you how your back will feel after four weeks.

The short version: start at medium-firm, move softer or firmer for your weight and position, and buy only with a long home trial.

How to choose a mattress: the five questions

Answer these before you look at a single product page:

1. How much do you weigh, and how are you built? Heavier and more muscular bodies sink deeper.
2. How do you sleep? Side, back, stomach, or a mix.
3. Does anything hurt? Low back, neck, shoulder or hip pain changes the priorities.
4. Do you sleep hot? Heat changes which materials suit you.
5. Do you share the bed? Motion and weight differences matter.

If you want the answers turned into a recommendation, the Mattress Lab runs these questions for you and shows why it suggests what it does.

What the evidence says about firmness

Medium-firm beat firm in the best trial

The strongest study on mattress firmness is a randomized, double-blind trial in The Lancet ¹. It enrolled 313 adults with chronic low back pain who had pain in bed and on rising. Half got a firm mattress and half a medium-firm one. After 90 days, the medium-firm group did better for pain while lying in bed and for disability. Pain on rising also leaned toward medium-firm, though that result alone was not conclusive.

That trial ended the old "sleep on a board" advice. It is still one trial, in one country, so the reviews matter too.

The reviews point the same way

• A 2015 systematic review of 24 controlled trials found a medium-firm mattress, or one sleepers could inflate and adjust themselves, best for comfort, sleep quality and spinal alignment ³.
• A 2026 review of 47 studies called medium-firm "the most defensible general recommendation" for adults with non-specific low back pain or morning stiffness ². It added that the choice should account for body build, age, sex-related body shape, sleep position and thermal comfort.

None of the reviews could pool their data, because studies used different mattresses and outcomes ². So medium-firm is a confident starting point. The exact number on the label is yours to find.

Why too soft and too hard both fail

A lab study measured the spine on soft, medium and hard mattresses ⁵. On the soft one, the head and neck dropped lower and computed peak loading on the spinal discs was 49% higher than on the medium one. On the hard one, the lower back curve flattened and contact pressure rose significantly. Medium sat in between. This was a small lab study, not a pain trial, but it fits the trial results.

The mattress firmness scale, decoded

Firmness labels look precise, but each brand sets its own.

• US 1-10 scale. 1 is softest and 10 firmest. A medium firm mattress is usually sold as about 6-7/10. Each brand rates its own products, so this is convention, not a standard.
• European H1-H5 grades. H2 and H3 are the common middle grades. Again, each manufacturer sets its own grades and weight ranges.
• The research scale runs backwards. The Lancet trial used a European hardness index where 1.0 was firmest and 10.0 softest ¹. Its medium-firm mattress scored 5.6. Do not compare that number with a 1-10 label in a shop.

Treat any label as a rough starting point. Your body on the mattress for a few weeks is the real test.

Match firmness to your weight and position

Body weight decides how far you sink. In a lab study of 59 adults, heavier people kept a more neutral spine on firmer mattresses, lighter and shorter people on softer ones, and people with wider hips deviated more on soft mattresses ⁷. The study was small, presented as a conference abstract and run with a mattress maker, so treat it as a guide.

Sleep position decides where you need give. Adults spend about 54% of their time in bed on their side, 38% on their back and 7% on their front, and side sleeping increases with age and body weight ⁶. A sagging surface hurt sleep mainly in side and stomach sleepers in a sleep-lab study ⁸.

The weight bands and numbers are our working guide built on the studies above. They are not research cut-offs. If you combine positions through the night, aim for the middle of your row and pick a responsive material that makes turning easy.

Side sleepers have the most to gain from fit. For more detail, see the best mattress for side sleepers and match your pillow height with the best pillow for side sleepers.

Choosing a mattress for back pain

If your back is the reason you are shopping, medium-firm is the evidence-based choice ¹²³. Adjust for weight: heavier people usually need the firmer end. Back sleepers can add a small pillow under the knees, and side sleepers a pillow between the knees.

Give it time. When healthy adults moved from their own beds, averaging 9.5 years old, to new medium-firm beds, back pain and stiffness fell and sleep quality improved, and the gains grew over four weeks ⁴. People on the cheapest category of old beds had more low back pain. These studies had no blinded control group, so part of the effect may be expectation. The full guide is in the best mattress for back pain.

Materials: what each one is good at

No trial shows one material is best for pain or sleep. The reviews point to firmness and fit ²³. Material changes how the bed feels, how warm it sleeps and how much it moves.

These traits come from industry convention and common experience. No trial has compared them head to head.

Temperature

Heat in bed fragments sleep. The 2026 review lists thermal comfort as one of the factors to weigh ². If you wake hot, favor pocket springs or latex, a breathable cover, and cotton or linen bedding before you pay extra for cooling technology.

Sharing the bed

Partner movement matters. In a lab study, a moving weight on one side of a mattress lightened deep sleep on the other side, and the effect was larger on mattresses that transferred more motion ⁹. If one of you is a light sleeper, choose good motion isolation. If your weights differ a lot, look at split-firmness mattresses or two single mattresses on one base.

Check the base, edges and height

A good mattress on a bad base is a bad bed. Foam mattresses usually need slats no more than about 7.5 cm (3 in) apart, and many warranties are void on wider gaps or a sagging base. That is warranty convention, so read yours. Compare options in bed frame vs base.

If you have nighttime reflux or snore, a head-up angle can help, and only some mattresses flex on a motorized base. See adjustable bed benefits. For older adults, firm edges and a bed high enough to stand up from easily matter as much as the mattress itself.

Test it at home

Bodies react differently, and showroom minutes tell you little. Prefer a mattress with a home trial of at least 100 nights, and read the return terms for pickup fees and minimum keep periods. Then:

1. Sleep on it for at least three to four weeks before deciding ⁴.
2. Rate your back, shoulders and sleep each morning on a 1-10 scale.
3. If you wake with shoulder or hip pain, it is probably too firm. If you wake with low back pain, it may be too soft.
4. Try a topper before you return a mattress that is close to right.

Replace your mattress if it has a visible dip, you can feel springs, or you wake stiffer than you went to bed. Bed age around 9.5 years was typical in the new-bed studies ⁴; the common 7-10 year guide is industry convention.

For people who train

Elite athletes often sleep under seven hours a night and sleep poorly, and an expert consensus recommends individual sleep strategies over one-size-fits-all rules ¹⁰. The mattress is one of the few parts of that you set once. Muscle is dense, so a muscular 90 kg (198 lb) athlete usually needs the firmer side of their weight band. No mattress has been shown to speed recovery, so buy for fit and temperature, and change beds in an easy training block, never in race week.

Common mistakes

• Buying the firmest option for a bad back. Medium-firm did better than firm ¹.
• Trusting a reviewer's body. A tester at 70 kg (154 lb) may love a bed that sags under you at 100 kg (220 lb).
• Reading H3 or 7/10 as a standard. It is one brand's label.
• Judging after one night. Use the trial.
• Ignoring the base. It can void your warranty and make the mattress feel softer.

When to see a doctor

A mattress can ease discomfort. It does not treat injury or disease. See a doctor if back or neck pain follows a fall, comes with fever, weight loss, numbness, leg weakness or bladder changes, or keeps you awake most nights for more than a few weeks.

Frequently asked questions

What firmness should I choose for my mattress?

Start with medium-firm, about 6-7 on a 1-10 scale where 10 is firmest. It did better than a firm mattress for back pain in a double-blind trial, and reviews of dozens of studies favor it for comfort and spinal alignment. Go a step softer if you are under about 60 kg (132 lb) or sleep on your side, and a step firmer if you are over about 90 kg (198 lb) or sleep on your stomach.

Is a firm mattress better for back pain?

Usually no. In the best trial, 313 adults with chronic low back pain had less pain in bed and less disability after 90 days on a medium-firm mattress than on a firm one. Very firm surfaces also raise contact pressure at the shoulders and hips. Heavier people may need the firmer end of medium-firm, but a rock-hard mattress is rarely the answer.

What does H2, H3 or H4 mean on a mattress?

These are European firmness grades, from H1 (soft) to H5 (very firm). Each manufacturer sets its own grades and weight ranges, so an H3 from one brand may feel like an H2 from another. The European test standard EN 1957 states that a firmness rating says nothing about comfort or quality. Use the grade as a rough starting point and test the mattress at home.

How long does it take to get used to a new mattress?

Plan on three to four weeks. In studies where people moved to new medium-firm beds, back comfort and sleep quality kept improving across all four weeks, so the first few nights are not a fair test. Rate how you feel each morning, and use the home trial period before you decide to return or exchange it.

How often should you replace a mattress?

The common guide of 7 to 10 years is industry convention. A better signal is how the mattress behaves: replace it if it has a visible dip, you can feel springs, or you wake stiffer than you went to bed. Springs in the middle of used mattresses were measurably weaker than at the edges, even when nothing looked wrong.

Is memory foam or a spring mattress better?

Neither is proven better for pain or sleep. Firmness and fit matter more than material. Memory foam gives close pressure relief and isolates motion well but can sleep warm. Pocket springs and hybrids breathe better and feel bouncier. Latex is springy and durable. Pick by how hot you sleep, whether you share the bed and how easily you need to turn.

Sources

1. Kovacs FM et al. (2003). Effect of firmness of mattress on chronic non-specific low-back pain: randomised, double-blind, controlled, multicentre trial. The Lancet.
2. Caggiari G et al. (2026). Mattress impact on lifestyle and back pain: a 25-year orthopedic literature review guideline. Journal of Orthopaedics and Traumatology.
3. Radwan A et al. (2015). Effect of different mattress designs on promoting sleep quality, pain reduction, and spinal alignment in adults with or without back pain; systematic review of controlled trials. Sleep Health.
4. Jacobson BH et al. (2008). Grouped comparisons of sleep quality for new and personal bedding systems. Applied Ergonomics.
5. Hong TT et al. (2022). The Influence of Mattress Stiffness on Spinal Curvature and Intervertebral Disc Stress: An Experimental and Computational Study. Biology (Basel).
6. Skarpsno ES et al. (2017). Sleep positions and nocturnal body movements based on free-living accelerometer recordings: association with demographics, lifestyle, and insomnia symptoms. Nature and Science of Sleep.
7. Shore H et al. (2019). Determining the ideal mattress firmness based on anthropometric measurements. Sleep Medicine (conference abstract).
8. Verhaert V et al. (2011). Ergonomics in bed design: the effect of spinal alignment on sleep parameters. Ergonomics.
9. Wellman JJ, Bohannon M, Vogel GW (1999). Influence of lateral motion transfer on sleep. Perceptual and Motor Skills.
10. Walsh NP et al. (2021). Sleep and the athlete: narrative review and 2021 expert consensus recommendations. British Journal of Sports Medicine.