Compound vs isolation exercises

You have been told that compound exercises build more muscle, and that isolation exercises are the finishing touch. It is the most repeated claim in strength training. Seven trials have tested it directly and it does not hold up. What the two do change is which lift gets stronger, and almost nobody tells you that part.

The short version

What the two words mean

A compound exercise moves more than one joint. A back squat bends your hip, your knee and your ankle together, so your quadriceps, glutes and hamstrings all take a share of the load. A bench press moves your shoulder and your elbow, so chest, shoulders and triceps split the work.

An isolation exercise moves one joint. A dumbbell curl bends your elbow. A lateral raise moves your shoulder. One muscle is doing the job it is named for, and the rest of you is mostly holding still.

That is the entire distinction. It describes the shape of the lift, not how hard it is, not how heavy it is, and not how useful it is. A heavy curl is still isolation. A light bodyweight squat is still compound.

This site uses those words on nearly every page in the library, and the muscle panel on each exercise page splits the work into what a lift drives hard and what it only steadies. That split is the same idea arriving from the other direction.

Which one builds more muscle

Neither, as far as anyone has been able to measure.

In 2023 a group led by Avery Rosa pooled every randomized trial that had compared the two head to head. Seven studies met the bar, giving ten separate comparisons. Their conclusion was that current evidence indicates similar whole muscle hypertrophy between multi-joint and single-joint training. Splitting the studies by whether training volume was matched between groups did not change the result.

The most pointed of the individual trials came from Paulo Gentil's group in 2013. They took 29 untrained young men and split them for ten weeks. One group did lat pulldowns and bench press. The other did lat pulldowns, bench press, and direct elbow work on top. Then they measured the arms by ultrasound.

The muscle grew 6.5% in the group with no arm work and 7.04% in the group that added it. Strength rose 10.40% and 12.85%. Neither difference was statistically significant. Adding curls and extensions to a program of two compound lifts produced, in the authors' own words, no additional benefits in muscle size or strength.

Read the limits with it, because they are real. Those men were untrained, they were young, they were all men, the study ran ten weeks and it measured one muscle group. It is one honest result, not a law.

The testosterone argument, and why it fails

Almost every page on this subject explains the supposed advantage the same way. Centr puts it as plainly as anyone: "The more muscles you engage, the more testosterone your body releases. More testosterone means you're capable of more muscle growth." No study is named alongside it.

The first sentence is probably true. The second is the one that was tested.

In 2012 Daniel West and Stuart Phillips took 56 young men through a 12-week program. Halfway through they measured the hormone surge after a hard leg session: growth hormone, free testosterone, IGF-1 and cortisol. Then they checked which men had gained the most muscle and strength. There were no significant correlations between the rise in growth hormone, testosterone or IGF-1 and gains in lean mass or leg press strength.

One awkward detail is worth keeping. Cortisol, the hormone usually cast as the villain, was weakly linked to gains in lean mass and in type II fiber size. Growth hormone tracked fiber size but not lean mass. Nothing tracked strength.

Robert Morton's group repeated the question in 2018 with 49 trained men and looked inside the muscle as well as in the blood. No hormone measured at any point was consistently linked to growth. The men who responded most and the men who responded least had no hormonal difference between them. What did track growth was how much androgen receptor the muscle carried, which is not something a choice of exercise changes on the day.

So the mechanism everyone repeats has been examined in 105 men across two studies and it did not predict who grew. That does not make compound lifts worse. It means the reason you were given for doing them was not the reason.

Which one builds more strength

Here the two genuinely differ, and the answer is simpler than the muscle question. You get strong at what you practice.

Nicolay Stien's group trained 53 physically active women for eight weeks. One group did leg press, a compound lift. One did leg extensions and kickbacks, two isolation lifts. A third did nothing. Everyone improved most at the exercises they had actually been doing. The isolation group gained more on leg extension and kickback than on leg press. The compound group gained more on leg press than on kickback.

The caveat matters as much as the finding. In the compound group, leg press and leg extension improved by similar amounts, and the gap did not reach significance. So training the compound lift carried over to the isolation lift reasonably well, while the reverse held less. The authors' own summary is that specific training is best, but that both approaches improved strength across all three exercises.

This is the useful half of the whole comparison. If you want a stronger squat, squat. If you want a stronger lateral raise, do lateral raises. Neither category is a shortcut to the other.

Whether compounds go first

The standard advice is to open your session with the big lifts and finish with the small ones. In 2021 João Nunes and colleagues pooled eleven studies that had tested exactly this.

Across all the strength tests together, order made no difference. Split by which kind of lift the strength was measured in, a clear pattern appeared: whichever exercise came first was the one that gained the most strength. Compound-first produced more compound strength. Isolation-first produced more isolation strength, and by a larger margin.

For muscle growth, order did nothing at all.

So "compounds first" is sound advice with the reason attached to the wrong outcome. It is not making you grow. It is deciding which lift you are prioritizing, which is worth doing on purpose rather than by habit. If your curl is the lift you care about this month, the research says put it first and accept a slightly weaker press that day.

The one place isolation may genuinely win

Rosa's review does not stop at "similar". It goes on to say that isolation work may allow you to preferentially grow subdivisions of a muscle, which could mean more complete development over time.

It then calls that hypothesis understudied and asks for more work on it. Both halves belong together. A muscle is not one uniform block, different exercises load different parts of it, and the idea that isolation work reaches parts a compound lift does not is plausible and thinly evidenced. Anyone telling you it is settled is going further than the paper.

What to actually do

Build the session around compound lifts, because they train more muscle per minute and that is a real advantage even though it is not a hormonal one. Then add isolation work for the specific things you want stronger or fuller, and accept that it is a preference rather than a requirement.

Three practical points fall out of the research above:

• Put the lift you care most about first. It is the one that will gain the most strength.
• If you are short of time, cutting isolation work costs you less than you have been told.
• Judge the decision on what got stronger, not on how the session felt.

That last one needs a record. The workout builder logs each set against a weight, so you can see whether the lift you put first actually moved.

Two other pages here already depend on this distinction. Mind-muscle connection is worth using on isolation work and not on heavy compounds. Training to failure suits the last set of an isolation exercise more than a heavy compound. And how many sets per exercise counts an indirect set as half a direct one, which is this same split expressed as volume.

Who this was measured in

Five of the seven sources on this page studied men only. Stien's trial is the one with women in it, 53 of them, and it is the specificity finding rather than the growth finding. Rosa and Nunes pool mixed sets of studies. That is the state of the evidence, and it is worth knowing before treating any of these numbers as universal.

Common questions

What is the difference between compound and isolation exercises?

A compound exercise moves more than one joint, so several muscles share the work. A squat bends the hip, knee and ankle. An isolation exercise moves one joint, so one muscle does the job. A bicep curl bends the elbow and nothing else. That is the whole distinction, and it describes the shape of the lift rather than how hard it is.

Do compound exercises build more muscle than isolation exercises?

No, and this is the claim most pages get wrong. Seven randomized trials have compared them directly. Pooled together they show similar whole-muscle growth. One trial added bicep and tricep work to a program of lat pulldowns and bench press for ten weeks and measured no extra arm growth at all.

Do compound exercises release more testosterone?

Probably, and it does not appear to matter. Two studies measured the hormone rise after training and then checked who actually grew. In 56 men, the rise in growth hormone, testosterone and IGF-1 did not predict gains in muscle or strength. In 49 more, no hormone measured in blood or muscle was consistently linked to growth.

Should you do compound exercises first?

If you want that lift to get stronger, yes. Pooling eleven studies, whichever exercise came first in the session was the one that gained most strength. That worked in both directions: putting isolation work first made the isolation lifts stronger. For muscle growth, the order made no measurable difference either way.

Can you build muscle with only compound exercises?

Yes. The trial that tested it gave one group lat pulldown and bench press only, and another group the same plus direct arm work. Both grew their elbow flexors about the same, roughly 7% over ten weeks. Those were untrained men and the study only measured the arms, so it is one honest answer rather than the final one.

Are isolation exercises a waste of time?

No. They win on two things the research does support. You get strongest at what you actually practice, so isolation work is how you strengthen a specific lift. And there is early evidence that isolation work may grow particular regions of a muscle rather than the whole thing, though the researchers who raise it call it understudied.

What are the best compound exercises?

The ones that move the most joints under the most load: squats, deadlifts, presses, rows, pull-ups and dips. Every one of them is in the library on this site with a clip. There is no ranking worth memorizing, because the research does not separate them by outcome.

Should beginners start with compound or isolation exercises?

Nothing here measured that directly. What the evidence does say is that a program of compound lifts alone produced the same arm growth as one with arm work added, in untrained men. So a beginner is not missing anything by starting with a handful of compound lifts and adding isolation work later.

Where these numbers come from

Seven papers, all in healthy adults, each read in full before anything here was written.

• Rosa A, Vazquez G, Grgic J, Balachandran AT, Orazem J, Schoenfeld BJ (2023). Hypertrophic effects of single- versus multi-joint exercise of the limb muscles: a systematic review and meta-analysis. Strength and Conditioning Journal 45(1):49-57. Seven trials, ten nested comparisons. The source of the "similar" and of the regional caveat. Its pooled estimate is described as trivial without being published, so no figure for it appears above.
• Gentil P, Soares SR, Pereira MC, da Cunha RR, Martorelli SS, Martorelli AS, Bottaro M (2013). Effect of adding single-joint exercises to a multi-joint exercise resistance-training program on strength and hypertrophy in untrained subjects. Applied Physiology, Nutrition, and Metabolism 38(3):341-4. PMID 23537028. The 6.5% against 7.04% arm growth.
• Brigatto FA, de Camargo JBB, de Ungaro WF, Germano MD, Marchetti PH, Aoki MS, Braz TV, Lopes CR (2020). Multi-joint vs. single-joint resistance exercises induce a similar strength increase in trained men. International Journal of Exercise Science 13(4):1677-1690. PMID 33414877. Sixteen trained men, six weeks each way with a washout between.
• Stien N, Pedersen H, Ravnøy AH, Andersen V, Saeterbakken AH (2020). Training specificity performing single-joint vs. multi-joint resistance exercises among physically active females. PLoS One 15(5):e0233540. PMID 32469946. The specificity finding, and the only source here with women in it.
• West DW, Phillips SM (2012). Associations of exercise-induced hormone profiles and gains in strength and hypertrophy in a large cohort after weight training. European Journal of Applied Physiology 112(7):2693-702. PMID 22105707. Fifty-six men, and the first half of the hormone answer.
• Morton RW, Sato K, Gallaugher MPB, Oikawa SY, McNicholas PD, Fujita S, Phillips SM (2018). Muscle androgen receptor content but not systemic hormones is associated with resistance training-induced skeletal muscle hypertrophy in healthy, young men. Frontiers in Physiology 9:1373. PMID 30356739. Forty-nine men, and the second half.
• Nunes JP, Grgic J, Cunha PM, Ribeiro AS, Schoenfeld BJ, de Salles BF, Cyrino ES (2021). What influence does resistance exercise order have on muscular strength gains and muscle hypertrophy? European Journal of Sport Science 21(2):149-157. PMID 32077380. Eleven studies, and the order finding.

What this page does not claim: that compound exercises build more muscle, that isolation work is unnecessary, or that testosterone does nothing. The pooled evidence does not support the first. Rosa's caveat about muscle regions cuts against the second. What was measured about hormones is narrower than the third: the rise after a session did not predict who grew.

If a lift hurts rather than being hard, stop and see a physiotherapist.