"Compound lifts are all you need" is one of the most repeated lines in gym folklore, usually followed by advice to drop curls, extensions, and lateral raises entirely. The controlled trials on this exact question tell a more specific story: compound and isolation exercises produce similar whole-muscle growth on average, compound exercises carry a real strength advantage for some lifts but not others, and isolation work can outright beat a compound exercise for growing one particular muscle if that muscle isn't the true limiting factor in the compound lift.
None of that is a reason to pick a side. It's a reason to know which exercise type answers which question, so a program isn't built on the version of the rule that happens to fit the trainee's existing habits.
What's the difference between compound and isolation exercises?
A compound exercise is a resistance exercise that moves two or more joints at once to produce force, such as the squat, bench press, or row. An isolation exercise is a resistance exercise that moves only one joint, such as a biceps curl, leg extension, or lateral raise.
The practical consequence of that difference is where fatigue shows up first. In a back squat, the hips, knees, and ankles all extend together, and whichever muscle group is weakest — often the lower back or hip extensors — ends the set before the quads reach true failure. In a leg extension, the knee is the only joint moving, so the quads can be driven to failure without the lower back or hips getting a vote.
Neither structure is better in the abstract. A compound lift trains several muscles per set and carries over to real-world and sporting strength; an isolation exercise trains one muscle per set and lets that muscle be pushed harder on its own terms. The research below is about when each structure actually produces a measurably different outcome.
Do compound exercises build as much muscle as isolation exercises?
Yes, on average — the best controlled comparison found no statistically significant difference in muscle growth between multi-joint and single-joint training over 10 weeks. Gentil, Soares & Bottaro (2015) put 29 untrained men through one of two programs: a multi-joint group training lat pulldown and bench press, and a single-joint group training biceps curl and triceps extension, both to the same volume and effort.
Muscle thickness, measured by ultrasound, increased 6.10% in the multi-joint group and 5.83% in the single-joint group — not a significant difference. Peak torque, a strength proxy, increased 10.40% in the multi-joint group and 11.87% in the single-joint group, again not significant. The authors concluded multi-joint and single-joint training were equally effective for upper-body size and strength in this population.
That result is specific to untrained lifters over a relatively short block, and it doesn't mean exercise selection is irrelevant — it means that, on average across a program, the joint category alone isn't what decides whether a muscle grows. What decides it is closer to the exercise itself, which the next two studies make concrete.
Do compound exercises build more strength than isolation exercises?
It depends on the lift, and the difference can be large or nearly zero within the same study. Brigatto et al. (2020) ran a crossover trial in 16 resistance-trained men, each completing six weeks on a multi-joint protocol (barbell bench press, incline bench press, incline barbell row, lat pulldown, back squat) and six weeks on a single-joint protocol (dumbbell flat fly, pec deck, straight-arm cable pulldown, dumbbell reverse fly, leg extension), separated by a three-week washout.
Every participant was tested on the same two lifts regardless of which protocol they'd just finished: barbell back squat 1RM and barbell bench press 1RM. For bench press, the multi-joint block produced a 10.8% 1RM increase (d = 0.60) versus 5.5% (d = 0.32) after the single-joint block — training the actual lift mattered for the test of that lift.
For squat, the gap nearly disappeared: 19.7% (d = 1.07) after multi-joint training versus 19.0% (d = 1.06) after single-joint training — despite the single-joint group never touching a squat rack, training their legs with leg extension alone. In these already-trained lifters, six weeks of pure leg-extension work produced almost the same squat 1RM improvement as six weeks of actually squatting.
For a self-coached lifter, the practical read is that specificity matters more for some lifts than others. If the goal is a bigger bench 1RM, benching is close to irreplaceable. If the goal is general lower-body strength, this trial suggests the leg muscles respond to loaded knee extension nearly as well as to the compound pattern itself — at least across a six-week block in trained men.
Can isolation exercises out-grow a compound exercise for the same muscle?
Yes — when the compound exercise you've chosen doesn't load that muscle as hard as a dedicated isolation exercise does. Mannarino et al. (2021) trained 10 untrained men for eight weeks in a within-subject design: one arm performed unilateral dumbbell rows (multi-joint), the other performed unilateral biceps curls (single-joint), for the same sets and reps.
The curl arm's elbow flexors grew 11.06% in muscle thickness. The row arm's elbow flexors grew 5.16% — less than half as much (p = 0.009). The row is a genuinely effective back exercise, but the biceps only act as a secondary mover in it; the lats and rear delts give out before the biceps come close to failure. The curl removes that ceiling entirely, since the elbow flexors are the only thing working.
"The single-joint exercise resulted in higher hypertrophy of the elbow flexors than multijoint exercise after 8 weeks of RT, whereas strength improvements were greater in accordance with specificity of RT exercise." — Mannarino et al. (2021)
The lesson isn't "stop rowing." It's that a compound lift only grows the muscles it actually pushes to fatigue, not every muscle technically involved in the movement. A lagging muscle that's a secondary mover somewhere in a program often needs its own dedicated exercise, not more volume on the compound lift it's hiding inside.
What lifters report in practice
Among self-coached lifters who track their training, the compound-vs-isolation debate tends to resolve the same way in practice as it does in the research: nobody who has trained for several years argues for dropping either category entirely. Lifters who train compounds almost exclusively commonly report specific muscles — rear delts, calves, biceps, triceps — stalling well before their squat, bench, or deadlift numbers do, which matches the mechanism in the Mannarino data above.
The more common criticism of isolation-heavy programming is time cost and recovery: adding several isolation exercises on top of a full compound-lift day is the most frequently reported reason lifters abandon an isolation addition within a few weeks, not that the exercises themselves didn't work. That's consistent with treating isolation work as a small, targeted addition rather than a program's foundation.
How do compound and isolation exercises compare at a glance?
| Attribute | Compound exercises | Isolation exercises |
|---|---|---|
| Whole-muscle hypertrophy | Similar to isolation on average (Gentil et al., 2015) | Similar to compound on average (Gentil et al., 2015) |
| Strength carryover | Strong for the specific lift trained (Brigatto et al., 2020) | Can match compound strength gains for some lifts, weaker for others |
| Targeting a lagging secondary mover | Limited — the prime mover usually fails first | Strong — the target muscle is the only one working |
| Time efficiency per session | High — several muscles trained per set | Lower — one muscle per exercise |
| Typical program role | The foundation of most sessions | A small, targeted addition for specific muscles |
How should self-coached lifters program and log compound vs isolation work?
Build most of a program around compound lifts, then add isolation exercises only where a specific muscle needs more direct stimulus than the compound lifts are giving it. That order matters for time and fatigue: compound lifts train more muscle per set, so they belong first in a session and first in the weekly structure, with isolation work layered on top rather than competing for the same recovery budget.
The harder part is knowing whether an isolation exercise you've added is actually working, and that's a logging problem more than a training problem. IronLedger tracks every exercise — a barbell row and a biceps curl included — as its own line with an independent personal record and Epley-based estimated 1RM, so a stalled curl trend line next to a rising row trend line is a visible, specific signal that a targeted muscle isn't being reached by the compound lift alone, exactly the pattern Mannarino et al. measured directly.
The progress page turns that comparison into per-lift trend lines and tonnage over time, and IronLedger's ten built-in program templates plus a custom multi-week builder let you schedule isolation exercises into specific days — after a push-pull-legs or upper/lower split's main lifts, for instance — without needing a separate spreadsheet to track which accessory work happened when. Pairing that with a sensible set count per muscle group keeps the isolation additions targeted instead of turning into unmanaged extra volume.
What mistakes do lifters make with compound vs isolation programming?
Most compound-vs-isolation mistakes come from treating the debate as a single either/or choice instead of a per-muscle decision.
- Dropping isolation work entirely because "compounds are enough." The Mannarino data shows a muscle that's a secondary mover in every compound lift in a program can lag for months without any dedicated exercise pushing it to failure.
- Judging an isolation exercise by soreness instead of trend data. Soreness the next day says nothing about whether a lift's estimated 1RM or working weight is actually moving; check the progress page trend line before deciding an exercise "isn't working."
- Assuming isolation results generalise across all lifts. Brigatto et al. found single-joint training matched compound training for squat 1RM but fell well short for bench 1RM — the size of the gap is lift-specific, not a fixed rule.
- Using sloppy form because the load feels light. Isolation exercises rely on strict joint mechanics to keep tension on the target muscle; swinging a curl or bouncing a leg extension shifts the work onto momentum instead of the muscle you added the exercise for.
- Adding isolation work without a plan for how many exercises per muscle is enough. See how many exercises per muscle group before stacking three or four isolation movements onto an already full session.
- Picking the wrong compound lift as evidence a muscle "doesn't respond" to it. A row that doesn't grow the biceps isn't proof the biceps grow slowly — it's a mismatch between the exercise chosen and the muscle being targeted, per the mechanism Mannarino et al. describe above.
Frequently asked questions
No. Mannarino et al. (2021) found single-joint curls produced more than double the elbow-flexor growth of multi-joint rows, because the row's back muscles fatigue before the biceps do.
Compound and isolation exercises answer different questions, not competing ones. Compound lifts are the more time-efficient foundation and the more specific test of the strength that actually matters to most goals; isolation exercises are the tool for a muscle that a chosen compound lift isn't reaching hard enough to grow. Log both as separate lines, watch the trend, and let a stalled number — not a rule of thumb — decide when to add one.
Sources and references
- Gentil, Soares & Bottaro. Single vs. Multi-Joint Resistance Exercises: Effects on Muscle Strength and Hypertrophy — 29 untrained men, 10 weeks, finding no significant difference in muscle thickness (6.10% vs 5.83%) or peak torque (10.40% vs 11.87%) between multi-joint and single-joint training (2015).
- Brigatto et al. Multi-joint vs. Single-joint Resistance Exercises Induce a Similar Strength Increase in Trained Men: A Randomized Longitudinal Crossover Study — 16 trained men, crossover design, finding multi-joint training clearly favoured for bench press 1RM (10.8% vs 5.5%) but statistically equivalent to single-joint training for squat 1RM (19.7% vs 19.0%) (2020).
- Mannarino et al. Single-Joint Exercise Results in Higher Hypertrophy of Elbow Flexors Than Multijoint Exercise — 10 untrained men, within-subject design, finding single-joint biceps curls produced 11.06% elbow-flexor growth versus 5.16% for multi-joint dumbbell rows (p = 0.009) (2021).
- IronLedger product page — verifies current per-exercise logging, independent personal records, and Epley-based estimated 1RM tracking (2026).
Keep learning
