The percentage is not the point: pH, free acid value, and what actually makes glycolic acid work

Two body lotions sit side by side. Both print the same glycolic acid percentage on the label. One transforms rough, bumpy skin within weeks; the other does almost nothing at all. The number that explains the difference is not on either bottle, and it is the only number that truly matters.

The percentage on a glycolic acid label sets only the ceiling of what is possible; how much of that acid is actually active depends on the formula's pH, and the resulting figure, the free acid value, is the real measure of strength. A buffered 12% glycolic acid body lotion at a sensible pH can deliver more usable, better-tolerated exfoliation than a higher percentage left raw and unbuffered.

This is the single most useful idea in body exfoliation, and almost nobody selling it will tell you, because it complicates a marketing story built entirely on one big number. Here is the chemistry, in plain language, and what it should change about how you read a label.

The number everyone reads, and the two they don't

Percentage is the headline. It is easy to print, easy to compare, and easy to misunderstand. A buyer scanning a shelf reasonably assumes that 15 beats 12 beats 10, the way a higher number beats a lower one everywhere else in life. For glycolic acid, that instinct is not just incomplete; it can be exactly backwards.

The percentage tells you how much acid is in the bottle. It does not tell you how much of that acid is in the form that can actually do anything to your skin. That second figure is governed by pH, and the relationship between the two produces a third number, the free acid value, which is what a serious standard measures rather than the headline alone. A meaningful judgement of a body lotion begins there, with how concentration and pH interact, not with the figure on the front of the label.

What free acid value actually means

Glycolic acid exists on the skin in two states. In one, the molecule holds onto its hydrogen and stays in its nonionic, free acid form, the form small and uncharged enough to slip into the skin and loosen the bonds between dead surface cells. In the other, it has given that hydrogen up, becoming a charged salt that penetrates poorly and does comparatively little. The proportion sitting in the active free form, at any given moment, is the free acid value.

What decides the split is pH measured against a fixed property of the molecule called its pKa, which for glycolic acid sits at around 3.83.¹ At a pH equal to the pKa, the molecules divide roughly in half, active and inactive. Crucially, every single point you drop below it represents a tenfold rise in acidity, tipping the balance steeply toward the active form. Researchers studying glycolic acid's antibacterial behaviour confirmed precisely this: the nonionic free form is the more active one, and potency climbed as pH fell across the tested range.¹ Strength, in other words, is not a quantity you pour in. It is a state you create with pH.

Why two identical percentages behave nothing alike

Picture two lotions, each genuinely 12% glycolic acid. The first is left at a very low pH, with almost all of its acid in the active form. It is potent, fast, and for body skin used nightly, potentially harsh. The second is brought up to a gentler pH, where a smaller but still meaningful fraction is active. It works steadily, comfortably, and can be used consistently, which is the only way exfoliation ever delivers.

Same percentage. Entirely different products. This is not theory: in human skin explant studies, glycolic acid formulated across a range of cosmetic concentrations and brought to a moderate pH still measurably stimulated epidermal renewal, while pushing the pH too high quietened the effect.² The percentage was constant. The behaviour followed the pH. A label that prints the first number and hides the second is showing you the part that flatters it.

The quiet case for buffering

If a lower pH means more active acid, the obvious question is why any thoughtful formulator would raise it at all. The answer is the difference between a product that works once and a product you can keep using.

Buffering means adding an agent that lifts the pH, deliberately trading a portion of raw potency for tolerability. It moves the formula closer to the skin's own slightly acidic surface, reducing the sting, the redness, and the barrier disruption that drive people to abandon an acid after a fortnight. The same explant research found that bringing glycolic acid to a moderate pH preserved its renewing effect without provoking the inflammatory signalling that harsher, lower-pH formulas can trigger.² The keratolytic action that smooths rough texture, the loosening of the cohesion holding dead cells in place, is well established as glycolic acid's central mechanism;³⁴ buffering simply lets that mechanism run at a pace skin can live with night after night.

There is a line past which buffering becomes self-defeating. A formula neutralised all the way to skin-neutral pH has converted almost all of its acid to the inactive salt and does very little, however impressive its percentage. The art sits in the middle: acidic enough to keep a real free acid value, gentle enough to use without flinching. We set out the wider version of this balance in our complete guide to glycolic acid body treatments and the standards a formula should meet.

What this means for a body, specifically

Body skin is thicker and more tolerant than the face, which is why it can carry an effective percentage at a sensible pH and resolve genuine texture: the keratosis pilaris on the upper arms, the strawberry legs, the rough patches that moisturiser alone never touches. The mechanism behind these conditions, a build-up of dead cells and keratin around the follicle, is exactly what a correctly bioavailable acid is built to address, as we explain in our guide to treating keratosis pilaris on the body.

But potency without comfort fails on the body for the same reason it fails on the face: an acid you cannot tolerate is an acid you stop using, and intermittent exfoliation produces intermittent results. This is why the supporting formula matters as much as the acid. Urea assists hydration and softening, niacinamide supports the barrier and calms visible redness, and shea butter restores the lipids exfoliation can strip. A six-marker standard treats free acid value, buffering and barrier support as inseparable, because a number that looks commanding on a label means little if the skin cannot stay the course. The relationship between strength and consistency is one we have explored in our examination of how a leave-on glycolic acid routine builds results over time.

The three numbers to ask for

If percentage alone is the wrong question, the right one is barely harder. To judge any glycolic acid product, you want three pieces of information rather than one: the total concentration, the pH, and whether the formula is buffered. Together they tell you the free acid value, and the free acid value tells you what the product will actually do.

A buffered formula in the region of pH 3.6 to 4.0 keeps a real, working fraction of acid active while staying tolerable enough for the regular use that texture change demands. A formula that prints only its percentage, and stays silent on pH and buffering, is asking to be judged on the one number it can win on. The category competes on that single figure precisely because the framework it omits is the one a well-made lotion already satisfies.

The quiet conclusion

Strength was never a number you could read off a bottle. It is the product of concentration and pH, expressed as the fraction of acid that is genuinely active and held at a level the skin can tolerate long enough to see results. A higher percentage at the wrong pH can be both harsher and less useful than a buffered 12% formulated with intent.

Once you know to ask for three numbers instead of one, the shelf rearranges itself. The loudest label is rarely the most thoughtful formula, and the question that sells the most product, how high is the percentage, turns out to be the wrong one entirely.