The $60 Billion Problem
You are standing in the supplement aisle, or scrolling a product page, holding something that promises to help you. Here is what the bottle will not tell you: no one outside the company that made it had to check it before it got there. The U.S. supplement industry sells over $60 billion a year, and the FDA does not approve any of it before it reaches the shelf. Manufacturers vouch for their own quality. Nobody has to prove a supplement works before selling it, and nobody has to prove the bottle actually holds what the label says.
Read that again. Prescription drugs survive years of clinical trials and FDA review before they reach you. Supplements do not - a 1994 law called DSHEA treats them more like food, so the FDA can usually only step in after a product has hurt someone. By then it is already in your cabinet.
That is how an honest product and a worthless one end up side by side on the same shelf, wearing nearly identical labels. Some hold exactly what they claim, in doses that actually did something in trials, checked by an independent lab. Others hold less than half the labeled amount, use a form your body barely absorbs, hide behind proprietary blends, and have never been tested by anyone outside the factory that made them. From where you are standing, the two look the same.
The review sites that should help you tell them apart are mostly making it worse. Search "best magnesium supplement" and you will find listicles ranking products the writers never tested, citing no research, earning a commission on every link. The scoring rubric, if there is one, got bolted on after the affiliate links were already in place.
We wrote our scoring methodology before we signed up for a single affiliate program. This page is the rest of it: how to think about whether a supplement works, and whether it was actually made well, so you can size up any product yourself - with our scores or without them.
The Evidence Hierarchy: Not All Research Is Equal
The supplement industry runs on phrases that sound like proof and aren't. "Clinically studied." "Research-backed." "Science-based formula." They are reassuring, and they tell you almost nothing, because none of them say what kind of research exists. One study in mice is research. So is a systematic review of 30 human trials. Treating those two as the same thing is exactly the move the marketing is counting on.
So here is the ranking, strongest evidence to weakest. If you take one thing from this page, make it this - it is the skill that protects you from everything else.
Systematic Reviews and Meta-Analyses
The strongest evidence there is. Researchers gather every relevant study on a question, weigh how good each one is, and combine the results - sometimes pooling the numbers mathematically, which is the meta-analysis itself. So when a meta-analysis of 15 randomized controlled trials finds that creatine monohydrate builds lean body mass, that is thousands of people across many research teams pointing the same way. You rarely get cleaner, and it is very hard to argue with.
Randomized Controlled Trials (RCTs)
The gold standard for a single study. People are randomly assigned to get either the real supplement or a dummy pill (a placebo), and ideally nobody - not the participants, not the researchers - knows who got which until the end (that is the double-blinding). It is the design that strips out wishful thinking and researcher bias. When we say vitamin D3 raises your blood level of vitamin D (your serum 25(OH)D), that comes mostly from RCTs.
But not all RCTs carry the same weight. A trial of 20 participants for 4 weeks tells you far less than one of 500 participants for 12 months. Sample size, length, who funded it, and whether the outcomes were locked in ahead of time all change how much the result is worth.
Cohort and Observational Studies
Here researchers follow groups of people over time and watch for links between taking a supplement and how their health turns out. These can surface a pattern, but they cannot prove the supplement caused it. The people who take vitamin D also tend to exercise more, eat better, and see a doctor regularly - so the link between vitamin D and better health might have nothing to do with the supplement at all. Observational studies are good at raising questions. RCTs are what answer them.
Animal and In-Vitro Studies
Studies in rodents, or in cells in a dish. They are useful for working out how something might work, but they fail to carry over to people all the time. A compound that shrinks a tumor in a petri dish can do nothing in a living person. One that sharpens memory in mice may never cross into a human brain at a realistic dose. When the marketing cites only animal studies, the honest translation is short: we do not know yet if this works in humans.
Traditional Use and Expert Opinion
The weakest evidence of all. "Used for centuries in traditional medicine" sounds like a credential and isn't one - bloodletting was used for centuries too. Expert opinion has its place in real clinical decisions, but it is not a stand-in for a controlled trial, and plenty of supplements the experts once endorsed fell apart the moment they were properly tested.
Why This Matters for Your Wallet
Most supplement marketing lives in the bottom two tiers. A company can truthfully call a product "backed by research" on the strength of a single mouse study. Knowing where a claim actually sits on this ladder is what keeps you from paying $40 a month for something that has never been shown to do anything in a human being.
Our Two-Tier Scoring System
Most supplement reviews jam two completely different questions into a single grade: does this ingredient work, and did this brand make the product well? That hides the very thing you need to see. A beautifully manufactured product full of an ingredient with zero clinical evidence is a different problem from a sloppily labeled bottle of something backed by 30 RCTs - and rolling them into one number buries the part you actually came for.
So we keep them apart. The full methodology has every detail; here is the shape of it.
Tier 1: Evidence Rating (per ingredient type)
Every supplement type gets an evidence tier based on how strong the clinical research is for its main claimed benefit. This is about the molecule, not the brand: whether you buy Thorne or Nature Made magnesium glycinate, the underlying clinical evidence is the same.
| Tier | Label | What It Means |
|---|---|---|
| 1 | Strong Evidence | Multiple large RCTs, systematic reviews, and meta-analyses confirm the benefit. The medical community broadly recognizes it. |
| 2 | Likely Effective | Several RCTs with generally consistent results. Good evidence base with some limitations in study design or sample size. |
| 3 | Mixed Evidence | Limited RCTs, inconsistent results, or primarily observational data. Reasonable people looking at the same research could disagree. |
| 4 | Weak Evidence | Mostly animal or in-vitro studies. Human data is scarce, from small trials, or poorly designed. |
| 5 | Ineffective | No meaningful evidence of benefit in humans, or evidence actively shows it does not work for its primary claimed purpose. |
This tier answers the first question: is the ingredient worth taking at all? If it lands at Tier 4 or Tier 5, it does not matter how beautifully a brand manufactured it.
Tier 2: Execution Score (per product, 0-100)
For ingredients that actually have evidence behind them (Tiers 1-3), we score each individual product on four equally weighted pillars, 25 points each.
Dosing & Form (0-25): Does the product give you the clinically effective dose identified in published RCTs, in a form your body can use? A magnesium oxide product scores lower than magnesium glycinate here for one reason: the research shows your body takes them up very differently.
Purity Verification (0-25): Has an outside lab actually checked this product - USP, NSF, ConsumerLab, BSCG, or Informed Choice? "Made in a GMP facility" and "this specific bottle had its contents verified by an independent lab" are not the same promise, and the gap between them matters.
Value (0-25): Cost per clinically effective daily dose. Not cost per pill, not cost per serving. If a product contains 150mg of an ingredient whose clinical dose is 600mg, you need four servings per day - so the real cost is four times what the label math suggests. This is the single metric most review sites get wrong.
Transparency (0-25): Does the label tell you every ingredient and its exact amount, and which form it uses? Or does it disappear into a "Proprietary Blend" that lists five ingredients and gives you only the total weight?
A product scoring 85-100 is excellent across all four pillars. One scoring below 40 has serious problems in more than one place. The full score interpretation is on the methodology page.
Category-by-Category Evidence Summary
Here is an honest map of where the evidence stands across the 15 supplement categories we cover. No marketing. Some are full of ingredients with strong research behind them; others run mostly on hype, and it is worth knowing which is which before you spend. You can browse them all on the categories index.
| Category | Evidence Landscape | Key Takeaway |
|---|---|---|
| Vitamins & Minerals | Strong for deficiency correction (Tier 1-2). Weak for supplementation in well-nourished populations. | Vitamin D3 is the standout - most adults in northern latitudes are insufficient. Magnesium is close behind. Most healthy adults eating a varied diet do not need a multivitamin. |
| Omega Fatty Acids | Strong for triglyceride reduction at high doses (Tier 1). Mixed for general cardiovascular protection. | Fish oil at 2-4g EPA+DHA has solid evidence for triglycerides. The "take fish oil for heart health" narrative is more nuanced than most brands suggest. EPA-heavy formulas have the strongest data. |
| Protein & Amino Acids | Strong for creatine and protein supplementation (Tier 1). Variable for individual amino acids. | Creatine monohydrate is one of the most researched and effective supplements in existence. Protein supplements work but are unnecessary if dietary intake is adequate. Most BCAA supplements are redundant if you eat enough protein. |
| Herbal & Botanical | Highly variable. Ranges from Tier 2 (ashwagandha for stress) to Tier 4-5 for many popular herbs. | This category has the widest gap between marketing claims and evidence. A few botanicals have genuine research support. Many are sold on tradition and animal data. Check the evidence tier before buying anything here. |
| Probiotics & Gut Health | Strain-specific evidence is moderate (Tier 2-3). Generic "probiotic blend" evidence is weaker. | Probiotics are not interchangeable. Specific strains have evidence for specific conditions. A random multi-strain blend from Amazon is not the same as the strain used in clinical trials. Digestive enzymes have limited evidence outside diagnosed deficiencies. |
| Sleep & Relaxation | Moderate overall. Melatonin is Tier 1-2 for sleep onset. Others are Tier 2-3. | Low-dose melatonin (0.3-1mg) works for falling asleep faster. Most products wildly overdose it. Magnesium glycinate has moderate evidence. L-theanine and 5-HTP show early promise but need more trials. |
| Joint & Bone Health | Mixed. Glucosamine sulfate has moderate evidence (Tier 2-3). Many others are Tier 3-4. | Glucosamine sulfate (not HCl) has the best data, but effect sizes are modest. Calcium + D3 is well-supported for bone density in deficient populations. Collagen for joints is an active area of research with early positive signals. |
| Cognitive & Nootropics | Mostly Tier 3-4. A few exceptions for specific populations. | The nootropics category is heavy on promise and light on proof. Creatine for cognitive performance has emerging evidence. Caffeine + L-theanine is well-studied. Most "brain stack" formulas combine underdosed ingredients with zero evidence for the specific combination. |
| Energy & Performance | Strong for caffeine and creatine (Tier 1). Variable for others. | Caffeine and creatine are proven performance enhancers. Beetroot (nitrates) has moderate evidence for endurance. CoQ10 is well-supported for specific populations (statin users, heart failure). Most pre-workout formulas underdose the ingredients that actually work. |
| Heart & Cardiovascular | Strong for omega-3 at prescription doses (Tier 1-2). Mixed for most OTC heart supplements. | High-dose EPA (prescription Vascepa-level) has cardiovascular event data from large trials. OTC fish oil at typical doses has weaker evidence. CoQ10 is well-supported for heart failure. Garlic extract has modest blood pressure data. Red yeast rice works but carries statin-like risks. |
| Weight Management | Mostly Tier 3-5. This is the most overhyped category in the industry. | No legal supplement produces meaningful weight loss without diet and exercise changes. Green tea extract has a small thermogenic effect. Fiber supplements support satiety. Everything marketed as a "fat burner" should be treated with extreme skepticism. |
| Immune Support | Moderate for deficiency correction (Tier 1-2). Weak for immune "boosting" in healthy people. | Vitamin D, zinc, and vitamin C support immune function when you are deficient. The evidence for "boosting" an already-healthy immune system is thin. Elderberry has some cold-duration data but the studies are small. The concept of "immune boosting" is itself misleading. |
| Skin, Hair & Nails | Limited. Tier 2-3 for collagen, Tier 3-4 for most others. | Collagen peptides have the most promising data for skin elasticity and hydration, though the studies tend to be small and often industry-funded. Biotin is effective for deficiency (rare) but has weak evidence for hair growth in people who are not deficient. |
| Men's Health | Mixed. Tier 2-3 for saw palmetto (BPH symptoms). Tier 3-4 for testosterone-focused supplements. | Saw palmetto has moderate evidence for urinary symptoms from BPH. Supplements marketed for testosterone "boosting" - tongkat ali, fenugreek, tribulus - have limited and inconsistent human data. Zinc matters if you are deficient. Most men's health supplements overpromise. |
| Women's Health | Strong for prenatal folate (Tier 1). Variable for others. | Prenatal folate for neural tube defect prevention is one of the best-established supplement benefits in all of medicine. Iron supplementation is critical for diagnosed deficiency. Most other women's health supplements fall in the Tier 3-4 range. |
Every supplement we cover has its own scorecard with the full evidence summary, product comparisons, and cost-per-dose math. Browse all 63 profiles on the supplements index.
What 90% of Supplement Buyers Get Wrong
Score enough products across enough categories and the same handful of mistakes keep surfacing. Here are the ones that cost people the most money.
1. Comparing price per bottle instead of cost per effective dose
A $15 bottle of magnesium oxide looks like the bargain next to a $30 bottle of magnesium glycinate. But your body absorbs only about 4% of magnesium oxide - you would have to swallow a pile of it to get what a single glycinate capsule delivers. Work out the cost per milligram you actually absorb and the "cheap" bottle is usually the expensive one. We run that math - cost per clinically effective daily dose - on every product we score, because it is the only price number that means anything.
2. Ignoring the form of the ingredient
Two products can list the same ingredient and not be the same product at all. Magnesium oxide vs. magnesium glycinate. Cyanocobalamin vs. methylcobalamin. Curcumin powder vs. curcumin with piperine or liposomal delivery. The form decides how much your body actually absorbs and uses, so a perfect dose of the wrong form can be nearly useless. Every one of our scorecards spells out which form the research supports.
3. Trusting "clinically studied" on the label
The phrase has no legal definition. It can mean the ingredient appeared in one tiny pilot study with 12 participants and no placebo group. It can mean the brand funded its own study with a design guaranteed to come out flattering. When we assign an evidence tier, we read the full body of research - systematic reviews first, then individual RCTs, weighed for quality, sample size, and whether the people who ran them were independent of the manufacturer.
4. Assuming third-party testing is standard
It is not. Third-party testing means an independent laboratory has confirmed the product contains what the label claims and is free from contaminants like heavy metals, pesticides, and undeclared substances. Fewer than 1 in 5 supplements on Amazon carry any form of third-party verification. The ones that do (USP, NSF Certified for Sport, ConsumerLab approved) spent real money on quality assurance. The ones that do not are simply asking you to take their word for it. Our purity verification pillar scores exactly this.
5. Buying proprietary blends
A "Proprietary Blend" on a label means the company folded several ingredients into one listed amount without telling you how much of each is in there. A blend might read "Proprietary Cognitive Complex: 500mg" over a list of five ingredients - and the lead ingredient could be 490mg, or it could be 5mg. You have no way to tell. Blends like that protect the manufacturer's margins, not some secret formula. A brand that takes transparency seriously just lists each amount.
6. Supplementing without testing
The most common way to waste money here is to take something you do not actually need. Vitamin D is a big deal if you are deficient and pointless if your serum levels are already optimal. Iron is essential for iron-deficiency anemia and can do harm if your iron stores are already adequate. Before you start anything, it is worth asking your doctor for the relevant blood work - a $30 lab test can save you hundreds on supplements you were never going to benefit from.
7. Stacking supplements with overlapping ingredients
Plenty of people take a multivitamin, a B-complex, an energy formula, and a stress supplement without noticing that all four are stacked with B vitamins. The overlap means you are paying several times over for the same thing, and in some cases pushing past a safe upper limit. Before you add anything new, read its full ingredient list against everything you already take.
8. Expecting supplements to fix what lifestyle changes would fix better
Supplements are exactly that - supplementary. They fill gaps; they do not replace sleep, exercise, a varied diet, or medical treatment. The strongest evidence for most supplements shows up in populations with a specific deficiency or a diagnosed condition. If you are generally healthy and eating a reasonable diet, the list of supplements with clear evidence of benefit is short: vitamin D in northern climates, creatine for performance, and omega-3 if you do not eat fish regularly. That is close to the entire list.
How to Use This Site
We built Supplement Scored to be the resource we wished existed when we started researching this ourselves. Here is how to get the most out of it, depending on where you are starting from.
If you know which supplement you want
Go to the supplements index and find the ingredient. Each scorecard opens with the evidence tier - does this ingredient work? - and then scores 8-15 specific products on dosing, purity, value, and transparency. The "Top Pick" and "Best Value" badges follow the execution scores, not which brand pays the highest affiliate commission.
If you have a health goal but are not sure which supplement to take
Start with the goals index. Each goal page - sleep, energy, joint health, focus, and 14 others - lists the supplements that matter for it, ranked by how strong the evidence is, so you can see which ingredients are even worth comparing before you start shopping. Or take the Find My Supplement quiz, which does the same thing interactively: pick up to three goals, set your budget and quality filters, and it ranks the evidence-backed options for you.
If you want to browse by category
The categories index organizes all 15 supplement categories with descriptions and links to every scored supplement in each one. Good for wandering through what exists in an area like probiotics or herbal supplements before you have a specific pick in mind.
If you want to understand our scoring
The methodology page walks through all of it: how evidence tiers are assigned, how the four execution pillars are scored, how cost per effective dose is calculated, and where our data comes from. We publish it because the methodology is the product. If you disagree with a score, you can trace exactly how we arrived at it.
If you want to go deeper on the evidence
The blog takes specific topics further. When You Actually Need a Supplement sorts out who benefits from supplementation and who is just spending. Supplement Ingredients Ranked by Evidence is the companion to this page, ranking individual ingredients by evidence tier. Supplement Certifications Explained breaks down what NSF, USP, and Informed Sport actually mean (and which ones matter for your situation). And our protein powder guide shows you how to calculate cost per gram of real protein so you stop overpaying. Every supplement scorecard also links straight to the primary research on PubMed.
The Bottom Line
The supplement industry makes good decisions hard, almost by design. The rules are weak. The marketing is loud. The review sites are conflicted. And most people are comparing the wrong numbers before they even start.
But the underlying research is real, it is sitting in public, and it is not that hard to read once you know what you are looking at. Some supplements work - for specific purposes, in specific populations, at specific doses. Many do not. The difference between burning $100 a month and spending $20 on something that actually helps comes down to three questions: Is the evidence real? Is the dose right? Is the product what it claims to be?
That is what we score. Every supplement, every product, every dollar made to justify itself.