The most common dosage mistakes across the top 50 supplements.
Roughly 80% of people are either underdosing compounds that require a threshold to work, overdosing things that have a ceiling above which you get nothing extra, or confusing compound weight with elemental or active ingredient content. Some of these mistakes are harmless waste. A few are genuinely counterproductive.
- NMN: Underdosing for age is the primary mistake. Under 35, 250 to 500mg daily. Between 35 and 50, 500mg. Over 50, the human RCT data trends toward 750 to 1000mg for functional outcomes in muscle and metabolic endpoints. The second mistake is running NMN inside a blend where it is the third or fourth ingredient at 50 to 100mg per serving. At that dose you are paying for a label claim. Third: not pairing with TMG. NAD synthesis at meaningful doses consumes methyl groups. 500mg TMG daily minimum is the appropriate co-supplement.
- NR. Same age-dose logic as NMN. The studied range is 250 to 1000mg daily. The specific mistake here is running NMN and NR simultaneously as though they stack additively. They share the same downstream pathway. Pick one and dose it properly.
- TMG. The cosmetic dose problem. Many multis include 50 to 100mg TMG. The studied range for homocysteine reduction and methyl donor support is 500mg to 3g daily depending on the indication. At 85mg you are not moving any needle. Standalone TMG powder is inexpensive and allows proper dosing. Above 3g in some individuals, trimethylamine odor develops via the FMO3 pathway. Start at 500mg and adjust.
- Magnesium. The single most common mistake. The label says 500mg magnesium. What is actually delivered depends entirely on the form. Magnesium oxide is 60 percent elemental by weight but so poorly absorbed that effective delivery is under 4 percent. Magnesium glycinate is roughly 14 percent elemental but well absorbed. Magnesium malate is approximately 15 percent. Magnesium threonate is approximately 7 percent. A product labeled 500mg magnesium glycinate is delivering roughly 70mg elemental. The target for sleep, muscle function, and neurological support is 300 to 400mg elemental daily. Check your label for the word elemental. If it is not there, calculate from the form.
- Vitamin D3. Two mistakes. First, dosing without testing. 2000 IU produces a very different serum level depending on where you start. Test 25-hydroxyvitamin D and dose to the target of 50 to 70 ng/mL, not by the label recommendation. Second: taking D3 without K2. Vitamin D increases calcium absorption. K2 MK-7 directs that calcium to bone rather than soft tissue. MK-7 at 100 to 200mcg is the studied dose for osteocalcin carboxylation. If your combined D3/K2 product uses MK-4 at 100 to 200mcg, that is not the studied dose. MK-4 trials showing bone benefit used 45mg daily, not micrograms.
- Omega-3 Fish Oil. The most persistent label fraud in the category. The front of the bottle says 1000mg fish oil. The supplement facts panel says 300mg combined EPA plus DHA. You are eating fish oil filler. The number that matters is combined EPA plus DHA. For cardiovascular protection, anti-inflammatory effects, and cognitive support the evidence is built around 1000 to 2000mg combined EPA plus DHA daily. Check the actual EPA plus DHA content, not the total oil weight. Triglyceride form absorbs better than ethyl ester form.
- CoQ10. Two mistakes. Form and dose. Ubiquinone requires conversion to ubiquinol in the body. Under 40 this is efficient. Over 45, conversion declines and ubiquinol is the right default. On dose: for general cardiovascular support 100 to 200mg is reasonable. For migraine prevention the AAN evidence is at 100 to 300mg. For egg quality and male fertility the studied range is 200 to 600mg. A single 50mg softgel in a multivitamin is doing nothing meaningful for any specific application.
- Creatine. The most studied supplement in existence and still commonly underdosed or over-complicated. 5g daily of creatine monohydrate is the standard. Creatine ethyl ester, buffered creatine, and most branded variants have no evidence of superiority over monohydrate. The loading phase is optional. The dose mistake is the 2 to 3g in blended pre-workout products. At 2g daily saturation takes significantly longer and most people do not maintain it consistently enough to get there.
- Berberine. The most commonly underdosed functional supplement in the longevity space. The clinical evidence for glucose regulation and AMPK activation is built around 1000 to 1500mg daily in divided doses, typically 500mg two to three times with meals. Products listing 200 to 300mg berberine as a therapeutic dose are selling you a fraction of what the evidence requires. Note also that berberine is a meaningful CYP3A4 and CYP2D6 inhibitor and should be disclosed to prescribing physicians at therapeutic doses.
- Trans-Resveratrol. Form is the primary mistake. Cis-resveratrol is largely inactive. Trans-resveratrol is the biologically relevant isomer. Products that say resveratrol without specifying trans are often a mixed or predominantly cis product. The second issue is bioavailability. Standard resveratrol has a short half-life. Pairing with quercetin or taking with a fatty meal improves tissue delivery. Below 150mg trans-resveratrol you are likely below the threshold for meaningful sirtuin activation. The studied range is 150 to 1000mg daily.
- Quercetin. Form and bioavailability are the issues. Standard quercetin has poor oral bioavailability due to limited water solubility. Quercetin phytosome and quercetin with bromelain have meaningfully better bioavailability data. The senolytic evidence used 500mg twice daily in intermittent protocols, not 50mg daily in a multivitamin.
- Apigenin (My personal fav): Consistently underdosed in combination products at 25 to 50mg. Standalone apigenin at 50mg is the commonly used dose for GABA-A modulation and sleep applications. For CD38 inhibition supporting NAD preservation, standalone dosing is the right approach over a blend with undisclosed amounts.
- Fisetin. The dosing strategy mistake is the most consequential here. Using fisetin as a daily antioxidant at 100mg is a different protocol from its senolytic application. The senolytic evidence is built on intermittent high-dose protocols, approximately 20mg per kg body weight over two consecutive days per month. For a 70kg adult that is 1400mg over two days, once monthly. Daily low-dose fisetin is an antioxidant play. Intermittent high-dose fisetin is the senolytic play. Know which one you are trying to execute.
- Spermidine. Dose confusion is widespread. Many products list the weight of the wheat germ extract, not the actual spermidine content. The human cognitive trial used 1.2mg actual spermidine daily from a standardized extract. Products claiming 10mg or 50mg are usually listing extract weight. Check the label for milligrams of actual spermidine, not extract weight.
- Urolithin A. The pomegranate extract mistake. Urolithin A is a gut metabolite and not everyone converts pomegranate ellagitannins to urolithin A. Gut microbiome composition determines conversion efficiency and a meaningful percentage of people produce little to none. Only pure urolithin A supplements guarantee delivery. The Mitopure trials used 500 to 1000mg daily of pure urolithin A.
- CA-AKG. The Rejuvant longevity trial used 1000mg daily and showed reduction in biological age markers over seven months. Common combination products include CA-AKG as a secondary ingredient at 300 to 500mg. If the goal is the epigenetic clock evidence, 1000mg is the dose the data is built on.
- Sulforaphane. Stability is the defining issue. Sulforaphane itself degrades quickly. Correct approaches are either myrosinase-active broccoli sprout extract, where the enzyme converts glucoraphanin to sulforaphane in the gut, or a stabilized sulforaphane product paired with myrosinase. Products listing glucoraphanin without active myrosinase rely on gut bacteria for conversion, which is variable. The studied dose range for Nrf2 pathway activation is approximately 10 to 40mg sulforaphane equivalent daily.
- Hyaluronic Acid. Molecular weight is the underappreciated variable. High molecular weight HA above 1000 kDa is the appropriate form for joint lubrication and skin hydration. Products that do not specify molecular weight are leaving out the most relevant information. Oral HA studies showing skin hydration benefit used 80 to 240mg daily of high molecular weight HA. Many products provide 20 to 40mg with no specification.
- Collagen Peptides. two mistakes. Dose and timing. The studied dose for skin, hair, and nail outcomes is 2.5 to 10g hydrolyzed collagen daily. For tendon and joint support the Shaw and Baar protocol uses 15g with vitamin C taken 30 to 60 minutes before mechanical loading. The timing mistake for tendon goals is taking collagen post-workout or with dinner when no loading is planned. Vitamin C co-administration is required in the Baar protocol.
- Vitamin C. The pharmacokinetic mistake. Plasma saturation occurs at around 200mg per dose. Taking 1000mg all at once is less effective than 500mg twice daily for maintaining tissue levels. For collagen synthesis support and immune applications, split dosing provides more sustained plasma availability. Above 2g daily GI side effects increase and absorption efficiency drops.
- Vitamin B12. Form matters enormously. Cyanocobalamin is synthetic and requires conversion steps impaired in older adults and certain genetic variants. Methylcobalamin and adenosyl-B12 are the active coenzyme forms. For people with slow COMT variants, methylcobalamin can drive catecholamine accumulation and adenosyl or hydroxocobalamin are better tolerated. For Metformin or long-term PPI users, standard serum B12 misses functional deficiency. MMA and homocysteine are the functional markers. Sublingual absorption bypasses the intrinsic factor requirement that declines with age.
- Folate. Folic acid is the synthetic oxidized form requiring DHFR enzyme conversion to active methylfolate. MTHFR variants, present in roughly 40 to 60 percent of the population in at least heterozygous form, reduce this conversion efficiency. For anyone with MTHFR variants or elevated homocysteine, 5-MTHF is the correct form. Most cheap multivitamins use folic acid.
- Iron: The most commonly over-supplemented mineral. Iron without confirmed deficiency or low ferritin is unnecessary and potentially harmful due to oxidative stress from excess free iron. Alternate-day dosing improves absorption over daily dosing because daily iron suppresses hepcidin. Ferritin below 30 is deficiency. Above 100 requires no supplementation. Test before you supplement, always.
- Zinc: Long-term zinc above 40mg daily depletes copper via metallothionein competition. Most people supplementing 25 to 50mg zinc are not co-supplementing copper. Therapeutic ratio is approximately 10 to 15 to 1 zinc to copper. Zinc picolinate and zinc bisglycinate absorb better than zinc oxide. Take with food to avoid nausea.
- Selenium. Narrow therapeutic window. The difference between adequate intake at 55mcg, the upper tolerable limit at 400mcg, and chronic toxicity is smaller than with most minerals. Many multivitamins already provide 100 to 200mcg. Adding a standalone selenium supplement on top can push into toxicity range with prolonged use. Selenomethionine is better retained than selenite.
- Iodine. Both directions are common mistakes. Vegetarians and vegans avoiding seafood are frequently deficient at RDA levels of 150mcg. People with Hashimoto's can worsen autoimmune thyroid inflammation with high-dose iodine supplementation. Check thyroid antibody status before adding iodine above RDA levels.
- Rhodiola Rosea. Standardization is the issue. The active compounds are rosavins and salidroside at a 3 to 1 ratio. The studied ratio is 3 percent rosavins and 1 percent salidroside. Products listing standardized rhodiola without specifying both compounds are unverifiable. Dose range is 200 to 600mg daily. Take in the morning on an empty stomach. Rhodiola is mildly stimulating and causes insomnia if taken in the afternoon. Tolerance builds, cycle it.
- Ashwagandha. KSM-66 and Sensoril are the extracts with human trial evidence. Generic ashwagandha root powder at unstandardized doses is pharmacologically unpredictable. The studied dose for cortisol and stress endpoints is 300 to 600mg of standardized extract. Higher is not meaningfully better above this range. Stop at confirmation of pregnancy.
- Lion's Mane. Mycelium on grain versus fruiting body extract is the dominant market problem. Most Amazon products are mycelium grown on grain substrate, meaning the product is substantially grain starch. Fruiting body extract standardized to beta-glucan content is the correct specification. Hericenones, the neuroactive compounds, are found in the fruiting body. The dose in the Mori 2009 cognitive trial was 1000mg dried fruiting body powder three times daily for 16 weeks.
- L-Carnitine and Acetyl-L-Carnitine. Not interchangeable. L-carnitine is the peripheral form for fatty acid transport and male fertility applications. Acetyl-L-carnitine crosses the blood-brain barrier and is the relevant form for cognition. For fertility, 1 to 3g L-carnitine daily is the studied range. For cognitive support, 1 to 2g acetyl-L-carnitine. A blended 250mg of either is subtherapeutic for both applications.
- NAC. Application-specific dosing is ignored by most people. For general antioxidant support, 600mg once or twice daily. For PCOS outcomes, 1200 to 1800mg daily. For psychiatric applications, 2400mg daily in divided doses was used in addiction trials. For mucolytic respiratory applications, 600 to 1200mg twice daily. Most stacks include one 600mg capsule without specifying what the goal is.
- Alpha Lipoic Acid. R versus S isomer distinction is consistently ignored. R-ALA is the biologically active naturally occurring form. Most cheap products are racemic, 50/50 R and S. S-ALA may partially antagonize R-ALA at the receptor level. Studies showing meaningful effects used 300 to 600mg R-ALA or 600 to 1200mg racemic ALA. The 15 to 25mg in multivitamins is cosmetic. Take on an empty stomach. Refrigerate R-ALA products due to stability.
- PQQ. The studied dose is 20mg daily for cognitive endpoints. 10mg is the lower bound showing any signal. Products listing 5mg PQQ are below the threshold used in positive human trials.
- Phosphatidylserine. The studied dose is 300 to 400mg daily in divided doses. Most nootropic blends include 50 to 100mg. Soy-derived PS is the source used in most human trials. Sunflower PS has a thinner evidence base despite being increasingly common.
- Citicoline. One of the most evidence-backed single-ingredient cognitive supplements and consistently underdosed in blends. The Cognizin form has the best human trial data. Studied dose is 250 to 500mg daily. Products including 50 to 100mg citicoline in a nootropic blend are providing a label claim, not a therapeutic dose.
- Lutein and Zeaxanthin. The AREDS2 trial used 10mg lutein and 2mg zeaxanthin daily. Most eye health supplements are in this range. The mistake is taking these without fat. They are carotenoids and fat-soluble. Take with your largest fat-containing meal. Products at 1 to 2mg lutein are below the AREDS2 dose.
- Curcumin. Standard curcumin has oral bioavailability below 1 percent without an enhancer. Meriva, BCM-95, Longvida, and Theracurmin all show meaningfully better plasma levels than standard curcumin. Piperine at 20mg increases bioavailability by approximately 2000 percent but interacts with multiple drug metabolizing enzymes. The hepatotoxicity signal is more common with enhanced forms because they work better, delivering hepatically meaningful concentrations in susceptible individuals. Baseline and follow-up liver enzymes at 3 months are warranted for anyone on high-dose enhanced curcumin formulations.
- Boswellia. AKBA content is what matters. Extracts standardized to at least 30 percent AKBA are meaningfully different from standard boswellia resin. 5-Loxin standardized to 30 percent AKBA is the most studied form. Generic boswellia at 400mg with no AKBA specification may contain very little active compound. Studied dose is 100 to 500mg of the standardized extract.
- Astaxanthin: Natural versus synthetic is the key distinction. Natural from Haematococcus pluvialis is the predominantly (3S, 3S) stereoisomer. Synthetic is a mixed isomer profile. BioAstin and AstaReal are the most studied natural sources. Dose range is 4 to 12mg daily. For eye health and photoprotection, 6 to 12mg has the better evidence. Fat-soluble, take with food.
- Probiotics. CFU count is the least informative number on the label. Strain specificity is what matters. Genus and species tell you almost nothing about clinical evidence. The strain designation, the letter-number suffix after the species name, is the clinically relevant identifier. Lactobacillus rhamnosus GG and Lactobacillus reuteri DSM 17938 have specific evidence for specific applications. A product listing Lactobacillus acidophilus at 10 billion CFU without a strain designation is clinically unverifiable.
- Glycine. The dose mistake is taking 500mg to 1g daily as a general wellness addition. The sleep quality evidence is built around 3g taken 30 to 60 minutes before bed. The GlyNAC protocol uses approximately 100mg per kg body weight daily, meaning 6 to 8g glycine for most adults. Below 3g the sleep-specific evidence is thin.
- Pregnenolone: Among the most frequently self-administered hormonal precursors and among the least predictable in conversion. Pregnenolone is upstream of DHEA, progesterone, cortisol, estrogen, and testosterone. Conversion is tissue-specific and individually variable. Common self-administration doses are 10 to 50mg daily. Monitoring DHEA-S, total and free testosterone, estradiol, and progesterone at baseline and after six to eight weeks is not optional. Taking this without periodic hormone testing is genuinely flying blind.
- DHEA. Gender-blind dosing is the main mistake. Women are meaningfully more sensitive than men due to lower baseline androgen levels. 25 to 50mg daily is the studied range in men. In women, 5 to 15mg is typically appropriate to avoid androgenic side effects including acne, hirsutism, and voice changes. Many products are sold at 50mg with no gender-specific guidance. Lab monitoring every six months is appropriate for both sexes.
- Melatonin: The pharmacological versus physiological dose distinction is the most important point here. The body produces approximately 0.1 to 0.3mg nightly. Most commercial melatonin is sold at 5 to 10mg, which is 20 to 100 times the physiological signal dose. Studies generally show no advantage above 0.5mg for sleep onset in most people and doses above 3mg can suppress endogenous production over time. For jet lag, 0.5 to 1mg is the studied dose. The exception is the oocyte protection data in fertility research which specifically used 3mg.
- Vitamin E. Most vitamin E supplements are dl-alpha-tocopherol, the synthetic racemic mixture. Natural vitamin E is d-alpha-tocopherol. More importantly, supplementing alpha-tocopherol alone at high doses displaces gamma-tocopherol, which has distinct anti-inflammatory properties. High-dose alpha-tocopherol above 400 IU daily has a concerning signal in the SELECT trial for prostate cancer and in meta-analyses for all-cause mortality at very high doses. Mixed tocopherol and tocotrienol products are a safer profile than isolated high-dose alpha-tocopherol.
- Biotin. The dose problem here is not efficacy but lab interference. The RDA is 30mcg. Most hair supplements contain 2500 to 10000mcg. Above 1000mcg, biotin interferes with multiple immunoassay-based laboratory tests including thyroid panels, troponin, FSH, LH, estradiol, and progesterone. Stop biotin for a minimum of 72 hours, ideally five to seven days, before any bloodwork. The evidence for biotin improving hair or nails in people without confirmed deficiency is weak.
- Taurine: The Singh et al. 2023 Science paper elevated taurine in the longevity conversation. The studied dose range in human trials is 1 to 6g daily. The common mistake is 500mg daily from a blend or energy drink, which is close to what food already provides. For longevity applications the animal data used doses equivalent to 3 to 6g daily in humans.
- Phosphatidylcholine. Dose for clinical applications including cognitive support and liver health is 1200 to 2400mg daily. Most combination products provide 200 to 400mg as a secondary ingredient. For vegetarians and vegans who avoid eggs and have low dietary choline, standalone phosphatidylcholine at a therapeutic dose matters more than a token amount in a blend.
- Myo-Inositol: For PCOS specifically, the evidence is at 2 to 4g daily. The 40:1 myo to D-chiro inositol ratio (as in Ovasitol) has stronger PCOS evidence than myo-inositol alone. Most products providing 500mg to 1g myo-inositol as part of a women's health blend are below the studied dose. For metabolic and fertility endpoints specifically, the dose matters.
- Vitamin A (Retinol): Frequently overlooked because it is in most multivitamins and cod liver oil simultaneously. Preformed retinol from all sources combined should stay under 3000mcg daily for adults and is particularly important in women planning pregnancy given teratogenic risk above this threshold. The mistake is not adding up the retinol across a multi, cod liver oil, and any other supplement containing preformed vitamin A. Beta-carotene does not carry the same risk as preformed retinol because conversion is regulated. Check the form.
So, the principle across all 50 is the same. Find the active ingredient dose, not the extract or compound weight. Confirm it matches what the human trials you are referencing actually used. Apply age and goal specific calibration. And stop testing multiple new supplements simultaneously because you will never know what is working.
Not medical advice though. Supplement decisions in the context of medical conditions or medications should be reviewed with the physician managing those conditions.
Let's discuss dosages.