Most pre-workout formulas contain three or four ingredients with a genuine human performance literature behind them, and then eight to twenty that are there to fill the label. The short answer to "what actually works" is unglamorous: caffeine, creatine (though not as an acute pre-workout ingredient), beta-alanine, dietary nitrate, and sodium bicarbonate are the compounds researchers keep returning to because there is measurable performance data to argue about. Almost everything else on a typical panel sits at the level of a plausible mechanism, a moved biomarker, or a sensation you can feel — which is not the same as an outcome.
What does "works" actually mean for a pre-workout?
This is where most marketing gets its leverage. There are four different things a supplement company can point to, and they are routinely presented as though they were interchangeable:
- Chemistry. The compound exists, dissolves, and is absorbed. Trivially true for most ingredients, and it proves nothing about training.
- A biomarker. Something measurable in blood or muscle changes — plasma nitrate rises, muscle carnosine rises, nitric oxide metabolites increase. This tells you the dose reached the target. It does not tell you a set went better.
- An intermediate outcome. Time to exhaustion, repetitions to failure, peak power, sprint time. This is the level at which ergogenic claims should be judged.
- A perceived effect. Tingling skin, a flushed face, a "pump", subjective alertness. These are sensations. Some correlate with an active dose; none is an outcome measure.
Almost every questionable pre-workout claim works by moving one level down this hierarchy and reporting it as though it were one level up. A vasodilation ingredient that raises a nitric oxide metabolite is described as improving performance. An ingredient that makes your skin prickle is described as "kicking in".
Which ingredients carry the strongest human performance evidence?
Ranked by how much direct human outcome data exists, not by how prominent they are on labels:
- Caffeine — the most heavily studied compound in the category. Its effects on perceived effort, alertness and endurance performance have been examined in humans repeatedly, across trained and untrained groups. This is the one ingredient where the burden of proof has been genuinely engaged rather than gestured at. It is also the ingredient responsible for most of what people feel from a pre-workout, which conveniently makes the rest of the blend seem to work.
- Creatine monohydrate — strong evidence, wrong category. The performance literature on creatine is large and the compound is cheap. But its mechanism is saturation of muscle phosphocreatine stores over days of consistent intake, so taking it in a pre-workout is a packaging decision, not a physiological one. Total daily intake matters; the pre-workout timing does not follow from the mechanism.
- Beta-alanine — supported, and slow. Beta-alanine raises muscle carnosine over weeks of daily intake, which is a buffering mechanism relevant to high-intensity efforts lasting on the order of a minute or two. Like creatine, it is a chronic-loading ingredient. The tingling sensation (paraesthesia) many people report is a dose-related nerve effect and not an index of efficacy.
- Dietary nitrate (beetroot) and sodium bicarbonate — real research areas, narrow use cases. Both have been studied as acute ergogenic aids in specific effort profiles. Both also appear on labels at token amounts that bear no relation to research protocols, and bicarbonate in particular carries a well-known gastrointestinal cost.
Names that get confused
Creatine is the active compound; creatinine is its breakdown product and a kidney function marker — different molecule, no ergogenic role. L-citrulline and citrulline malate are not the same label claim: the malate version is a mixture, so a gram figure for citrulline malate is not a gram figure for citrulline, and the ratio is often unstated. Beta-alanine is not L-alanine. Arginine AKG is a different compound from arginine hydrochloride, and neither inherits citrulline's data.
Where does the evidence get weak, and what are the real drawbacks?
Three problems recur, and none is exotic.
Underdosing behind a proprietary blend. When a label lists fifteen ingredients under a single blend weight, the arithmetic is doing the marketing. Ingredients with genuine research usually require gram-scale doses; a blend that totals a few grams cannot deliver research-scale amounts of more than one or two of them. The named ingredients are real; the doses often are not.
Caffeine doing the work and getting the credit. If a formula produces a noticeable acute effect, caffeine is the most likely explanation. That makes personal experience an unreliable guide to whether the other seventeen ingredients contribute anything, and it makes tolerance a genuine issue — the effect you are paying for is the one that fades with habitual use.
Comparisons nobody has run. Formula A beating placebo and formula B beating placebo does not establish that A beats B. I have not found direct head-to-head human trials for most of the specific blend-versus-blend, ratio-versus-ratio, or "advanced form" claims used to justify price differences. Those questions are unanswered, not settled in the expensive product's favour.
The practical costs are ordinary: overstimulation and disrupted sleep from late dosing, gastrointestinal discomfort from bicarbonate or from large sugar-alcohol loads, and paraesthesia from beta-alanine. Anyone managing blood pressure, cardiac issues, pregnancy, or stimulant-interacting medication should treat a high-stimulant blend as a pharmacological decision rather than a food choice, and clear it with a clinician.
How do the plausible-but-unproven ingredients propose to work?
The "pump" and "focus" categories run on mechanisms that are chemically coherent and clinically underdetermined.
The nitric oxide group — citrulline, arginine, nitrate sources — is built on a real biochemical pathway: these compounds feed nitric oxide synthesis or provide nitrate for reduction, and nitric oxide relaxes vascular smooth muscle. That is established chemistry. What it establishes is a vasodilation mechanism, not a training outcome. A visible pump is haemodynamics, not hypertrophy, and citrulline generally has a better absorption profile than oral arginine — which is a pharmacokinetic argument, not a performance one.
The cognitive ingredients — tyrosine, alpha-GPC, huperzine A, theanine, various "nootropic" additions — are proposed as precursors or modulators of neurotransmitter systems. A precursor raising a neurotransmitter pool is a mechanism at the exposure-biomarker level. Whether that translates into better performance in a training session, at the doses actually used in blends, is largely unknown rather than disproven. BCAAs and taurine sit in similar territory: defensible biochemistry, weak grounds for the specific claims printed beside them.
Then there are the ingredients that are pure decoration — trace amounts of exotic botanicals, vitamins added at levels that do nothing acutely, and niacin included at a dose that produces a flush people interpret as potency. The flush is a receptor-mediated vasodilation response. It is a sensation, not an ergogenic signal.
Who has something to gain — and who doesn't?
A stimulant pre-workout is most defensible for people doing genuinely hard training who are not caffeine-habituated and who train early enough that a stimulant will not compromise sleep. For them, caffeine has a real literature and the rest is optional.
It is least defensible for people already drinking coffee through the day (you are buying caffeine you already consume, with fifteen unpriced extras attached), for people training within several hours of bed, and for anyone whose sessions are moderate enough that acute ergogenic effects are not the limiting factor. If sleep, protein intake, and programme consistency are unresolved, no blend will compensate.
How should you rank what's on the label?
Judge by evidence tier, not by ingredient count:
- Fully disclosed doses per ingredient, no proprietary blend. Direct human outcome evidence is dose-dependent, so a hidden dose makes every other claim unverifiable. This is the only non-negotiable criterion.
- Caffeine content stated in milligrams. Direct human outcome evidence supports caffeine, and the number is what lets you manage tolerance, timing, and total daily intake.
- Research-scale amounts of beta-alanine, creatine, or nitrate if you want them. Biomarker-supported and outcome-relevant, but only at gram-scale intake taken consistently — and creatine and beta-alanine work through daily accumulation, which means a standalone daily dose is the sensible route regardless of what any pre-workout contains.
- Citrulline and the nitric-oxide group. Mechanistically plausible, largely unresolved at the outcome level. Reasonable to include, not a reason to pay a premium.
- Nootropic blends, exotic botanicals, "advanced" forms of studied compounds. Untested at the claim's level. A newer form is not automatically better; it is usually just less studied, and price is not an evidence tier.
- Third-party testing for banned substances. Not an efficacy criterion at all, but the one that matters if you are drug-tested.
The verdict
The honest version of a pre-workout is a short list: caffeine before training if it suits your schedule and tolerance, creatine monohydrate taken daily at any time, and beta-alanine daily if your sport involves repeated near-maximal efforts. Everything past that is a mechanism you are paying to hope about.
If you buy a formula, buy the cheapest fully disclosed one that hits the doses you actually want, and treat the ingredient count as a cost rather than a feature. If you can tolerate the less exciting option, buying caffeine and creatine separately gives you the same evidence base with the doses under your control — and removes the possibility that what you are feeling is a niacin flush and a plausible story.
