How to choose

What supplements should I take? A method, not a stack.

The wellness wave sells stacks without measuring you. Here is how to pick the few supplements with evidence for your goal, and test them on your own data.

By the Umbral team9 min read114 studies cited

Quick answer

The supplements you should take are the few with strong evidence for your one goal, started one at a time: omega-3 for resting heart rate (about 2 bpm lower), saffron for sleep quality (2.14 points better), lavender oil for stress (anxiety 2.8 points lower), creatine for strength (3–5 g a day). Test before iron, vitamin D or B12.

Key takeaways

  • A grade measures how solid the research is, not how big the effect is: MelatoninStrong is well studied and speeds sleep onset by only about 7 minutes.
  • Take ironStrong only after a ferritin test, never blind, and check your medication before starting anything new.
  • Measure one supplement at a time against your band (your normal range): Omega-3 (EPA+DHA)Strong lowers resting heart rate about 2.2 bpm, enough for a wearable to show.
In this guide

How do I choose the right supplements?

Choose supplements the way you would run a fair test: one goal, the best evidence for it, a blood test where one applies, a safety check, then one supplement at a time. Many people need fewer supplements than the wellness wave suggests, and some need none. Five steps find the few that fit you.

  1. 01Name one goal. Sleep, stress, energy or training. One goal means one metric to watch, and that is what makes a result readable.
  2. 02Choose by evidence grade. Start with Strong or Moderate picks; leave Emerging for later and read Caution as a reason to skip.
  3. 03Test before ironStrong, vitamin DModerate or B12Moderate. All three help only if you are low, and extra iron can harm you.
  4. 04Check interactions and safety. Blood thinners, thyroid medication, SSRIs, blood-pressure medication, sedatives, immunosuppressants, pregnancy and kidney disease all narrow your options.
  5. 05Start one at a time and measure. Umbral compares a 21-day trial, logged with a one-tap weekly check-in, against the 14-day baseline your wearable already holds. The verdict reads as association, not causation.

Which supplement is best for your goal?

The best supplement for you is the one with the strongest evidence for your single goal: saffronStrong or AshwagandhaStrong for sleep, Lavender oil (Silexan)Strong for stress, Creatine monohydrateStrong for strength and Beetroot nitrateStrong for endurance. Tiredness starts with a blood test, not a pill. The linked guides go deeper on each goal.

Where to start, by goal
GoalStart withEvidenceGuide
SleepSaffron, ashwagandha, Tart cherryModerateSleep quality 2.14 points better across 8 trialsSleep guide
StressLavender oil, L-theanineModerateAnxiety 2.8 points lower than placeboStress guide
TirednessIronStrong after a ferritin test; Panax ginsengModerateIron: tiredness down 48% vs 29% on placebo when ferritin was lowEnergy guide
StrengthCreatine, Whey proteinStrongWhey: +5.5 lb (2.49 kg) on your best liftTraining guide
EnduranceBeetroot nitrate, CaffeineStrong, Beta-alanineStrongBeetroot: time trials 1–3% fasterTraining guide
Workout recoveryCurcuminStrong, whey proteinCurcumin: soreness 0.56 lower on a 0–10 scaleTraining guide
Resting heart rate and HRVOmega-3 (EPA+DHA)Strong, TaurineModerateOmega-3: 2.2 bpm lower across 51 trials—
Cold seasonZinc lozengesModerate within 24 h of symptoms, vitamin CModerate daily, probioticsModerateZinc: colds about 2.4 days shorter, low certainty—
Premenstrual symptomsCalciumModerate, Vitex (chasteberry)Moderate, Magnesium + B6ModerateCalcium: symptoms down 48% vs 30% on placebo—

Which supplements have the best evidence?

Six supplements pair a Strong grade with an everyday goal: Omega-3 (EPA+DHA)Strong, AshwagandhaStrong, saffronStrong, Beetroot nitrateStrong, Creatine monohydrateStrong and Lavender oil (Silexan)Strong. They are ranked by grade, then by whether a wearable can see the effect. Strong means the research is solid; the effects themselves are still modest.

The best-evidenced shortlist, ranked
Supplement · EvidenceBest forWhat studies foundTypical doseOn your wearable
Omega-3 (EPA+DHA)
Strong
Resting heart rate and HRVResting heart rate about 2.2 bpm lower across 51 trials.1–3 g/day EPA+DHALikely visible
Ashwagandha
Strong
Stress-linked poor sleepCortisol 27.9% lower after 60 days at 600 mg.600 mg/dayLikely visible
Saffron (Crocus sativus extract)
Strong
Sleep quality, especially with low moodSleep-quality score 2.14 points better than placebo across 8 trials.28–30 mg/day standardized extractLikely visible
Beetroot nitrate
Strong
Same pace at a lower heart rateTime trials 1–3% faster; feeds your wearable's VO₂ max estimate.6–13 mmol nitrateLikely visible
Creatine monohydrate
Strong
Strength, power and short-sleep daysBest-studied performance supplement; one dose eased thinking deficits after 21 h awake.3–5 g/dayBelow wearable noise
Lavender oil (Silexan)
Strong
Anxious, restless eveningsAnxiety scale 2.8 points lower than placebo at 80 mg over 10 weeks.80 mg/dayBelow wearable noise

Omega-3 (EPA+DHA)

Strong

DHA-forward fish or algal oil

Suits almost anyone who wants a number to watch. The drop is driven by DHA and takes 4–12 weeks, and it is one of the few supplement effects steady enough to show in your own resting-heart-rate trend.

Dose
1–3 g/day EPA+DHA
Timing
With a fat-containing meal; 4–12 weeks to effect
Watch out for
  • · Mild bleeding-time increase — caution with anticoagulants; stop 2 weeks pre-surgery.
  • · Atrial fibrillation signal at ≥4 g/day — stay ≤3 g.

Ashwagandha

Strong

Root extract (KSM-66 / Shoden)

Suits stress-linked poor sleep, not people who already sleep well. Benefits build over 8 weeks or more; fewer night wakings and higher overnight HRV are changes a wearable can plausibly show.

Dose
600 mg/day (or 300 mg ×2)
Timing
Evening; benefits build over 8+ weeks. Cycle 8–12 weeks.
Watch out for
  • · Rare liver-injury case reports (NIH LiverTox) — stop on dark urine/jaundice.
  • · Raises thyroid hormones — avoid with hyperthyroidism or levothyroxine.
  • · Avoid in pregnancy and autoimmune disease.

What do evidence grades mean?

An evidence grade tells you how much research stands behind a supplement, not how large its effect is. Umbral uses three grades (Strong, Moderate and Emerging) and one flag, Caution, for a safety or interaction concern that outweighs the benefit. One real example of each:

  • Strong · multiple randomized trials or meta-analyses. MelatoninStrong: 19 trials agree on about 7 minutes faster sleep onset. Strong evidence, small effect.
  • Moderate · some trials, limited replication. MagnesiumModerate: about 17 minutes faster sleep onset, from 3 small trials in older adults with insomnia.
  • Emerging · preliminary or mechanistic. PhosphatidylserineEmerging: exercise cortisol about 35% lower in one study of 10 people.
  • Caution · interaction or safety flag. EchinaceaCaution: a modest effect at best, against documented allergic reactions including anaphylaxis.

Which supplements need a blood test first?

Three supplements need a blood test first: ironStrong, vitamin DModerate and vitamin B12Moderate. Each helps when you are actually low, does little or nothing when you are not, and iron can do harm. If you are asking which vitamins you should take, start here.

Test first, then decide
SupplementTestWhy test firstWho should ask
IronFerritin + CBCIn women with ferritin under 50 µg/L, tiredness fell 48% vs 29% on placebo. Hemochromatosis affects about 1 in 250 people; excess iron is pro-oxidant.Menstruating women, endurance athletes, vegetarians
Vitamin D25(OH)DHelps tiredness and weakness only if you are deficient. Chronic high doses raise blood calcium; interacts with thiazide diuretics. Retest in 3 months.Little sun, darker skin at high latitude
Vitamin B12Serum B12 (± MMA)No effect on tiredness or thinking when levels are normal.Vegans, adults over 50, metformin or PPI users

Which supplements interact with medication?

Twenty-five supplements in Umbral's catalog carry a flag for a medication class or a life stage: blood thinners, thyroid medication, SSRIs, blood-pressure medication, sedatives, immunosuppressants, pregnancy and kidney disease. Find your row in the table below before you start anything new.

Interaction flags by medication class
Medication or situationCheck before starting
Blood thinners (anticoagulants)Omega-3 (EPA+DHA)Strong, MelatoninStrong, CoQ10 / ubiquinolModerate, saffronStrong, Apigenin (chamomile-derived)Emerging, Panax ginsengModerate, CurcuminStrong, L-carnitineModerate, EchinaceaCaution
Thyroid medicationAshwagandhaStrong, Whey proteinStrong, L-carnitine, CalciumModerate
SSRI antidepressantsRhodiola roseaEmerging, saffron, Panax ginseng
Blood-pressure medicationL-theanineModerate, Beetroot nitrateStrong, Panax ginseng, TaurineModerate, Citrulline malateModerate
SedativesAshwagandha, apigenin, lemon balmModerate, Vitamin B6 (pyridoxine)Caution
ImmunosuppressantsAshwagandha, melatonin, probioticsModerate, echinacea
PregnancyAshwagandha, saffron, Lavender oil (Silexan)Strong, apigenin, lemon balm, Panax ginseng, curcumin, taurine, Sodium bicarbonateStrong, QuercetinEmerging
Kidney diseaseMagnesiumModerate, Creatine monohydrateStrong, curcumin, taurine, whey protein, sodium bicarbonate

Specifics: omega-3 slightly lengthens bleeding time, CoQ10 may weaken warfarin, ashwagandha raises thyroid hormones and melatonin also interacts with contraceptives. This is not a full interaction check: talk to a clinician before starting.

How do you buy a supplement you can trust?

Buy the form the trials tested, from a brand that proves what is in the bottle. Labels are not always accurate and some categories carry documented contamination, so third-party testing and a certificate of analysis matter more than price or packaging. Five cases from the catalog:

  • MelatoninStrong: measured content ran from 83% below to 478% above the label. Choose third-party-tested, USP-verified products.
  • Whey proteinStrong: look for Informed Sport or NSF certification.
  • CordycepsEmerging: wild Cordyceps sinensis is often adulterated, with documented lead and arsenic. Buy cultivated C. militaris or Cs-4, never a mushroom blend.
  • Citrulline malateModerate: check the 2:1 ratio on the certificate of analysis; many products deliver 1.1:1, about half the labeled citrulline.
  • Cocoa flavanolsModerate: cocoa can carry cadmium, so prefer brands with third-party heavy-metal testing.

What we left out and why

Some popular supplements are not in Umbral's recommendations, and two more carry a Caution flag. If the evidence is thin or the risk is real, it does not belong on your shelf.

  • Kava: reports of liver toxicity; banned or under warnings in several countries.
  • Valerian: people rated their sleep better, but actigraphy showed no objective change.
  • Oral GABA: doubtful that it crosses the blood–brain barrier; low-quality evidence.
  • EchinaceaCaution: nothing measurable once you are sick, and anaphylaxis and acute asthma are documented.
  • Vitamin B6 (pyridoxine)Caution: the doses that eased premenstrual symptoms (50–600 mg/day) sit above EFSA's 12 mg/day upper limit, and high doses over time can damage nerves.

How do you know if a supplement is working?

A supplement is working for you when your own metric shifts while you take it and moves beyond its usual night-to-night swing. That needs an effect big enough for a wearable to see, so check the tier first: visible, real but hidden, or out of reach.

Umbral keeps the test fair: a 14-day baseline your wearable already recorded, one supplement for 21 days with a one-tap weekly check-in, and a verdict after at least 7 nights on each side, read as association, not causation. Connect the wearable you already own, start one supplement, and measure it against your own band.

Frequently asked questions

How many supplements should I take at once?

One new supplement at a time. If you start two together and your sleep or resting heart rate changes, you cannot tell which one did it, or whether either did. Run one for 21 days against your baseline, read the verdict, then add the next.

Do I need a blood test before taking supplements?

Only for some. Iron needs a ferritin test with a complete blood count first, never blind, because hemochromatosis affects about 1 in 250 people. Vitamin D needs a 25(OH)D test and B12 a serum B12 test, since both help only when you are low.

How long does it take for a supplement to work?

It depends on the supplement. Caffeine is taken about 60 minutes before exercise. Omega-3 needs 4 to 12 weeks, ashwagandha 8 weeks or more, and vitex three menstrual cycles. Know the timing so you neither quit a slow supplement early nor credit a fast one for noise.

Is it safe to take supplements with medication?

Not always. Omega-3, saffron and CoQ10 are flagged with blood thinners, ashwagandha with thyroid medication, saffron and rhodiola with SSRIs, and beetroot nitrate and L-theanine with blood-pressure medication. Talk to a clinician before starting anything if you take medication, are pregnant or manage a condition.

Should I take supplements in the morning or at night?

Follow the supplement, not a rule. Rhodiola is mildly stimulating, so take it in the morning; Panax ginseng never in the evening, since insomnia is a documented side effect. Glycine goes 30 to 60 minutes before bed, magnesium 1 to 2 hours before. Creatine works at any hour.

Which supplements can a wearable actually measure?

Only those with effects bigger than your night-to-night noise. Omega-3 lowers resting heart rate about 2.2 bpm over 8 to 12 weeks, which a wearable trend can show; tart cherry and beetroot nitrate are also visible. Melatonin's 7 minutes faster sleep onset is real but hidden in nightly variation.

Find out which of these move your numbers.

A guide can rank the evidence; only your own data can say what works for you. Umbral learns your band (your normal range) for sleep, HRV and recovery, recommends the supplements with evidence for what it sees, and measures each one against you. Results read as association, not causation.

Just connect the wearable you already own: WHOOP, Oura, Garmin, Fitbit and more. Nothing new to buy, no blood draw to start.

References

Every number in this guide comes from these studies: the same evidence catalogue that grades recommendations inside Umbral.

Saffron (Crocus sativus extract)

  1. Effects of saffron supplementation on improving sleep quality: a meta-analysis of randomized controlled trials
    Sleep Medicine · 2022 · meta-analysis · 8 trials · n = 611
  2. A standardised saffron extract improves subjective and objective sleep quality in healthy older adults with sleep complaints: results from the gut-sleep-brain axis randomised, double-blind, placebo-controlled pilot study
    Food & Function · 2025 · pilot · n = 52
  3. An investigation into an evening intake of a saffron extract (affron®) on sleep quality, cortisol, and melatonin concentrations in adults with poor sleep: a randomised, double-blind, placebo-controlled, multi-dose study
    Sleep Medicine · 2021 · rct · n = 120
  4. Effect of a saffron extract on sleep quality in adults with moderate insomnia: a decentralized, randomized, double-blind, placebo-controlled trial
    medRxiv / Current Research in Food Science · 2025 · rct · n = 158
  5. Effect of saffron supplementation on symptoms of depression and anxiety: a systematic review and meta-analysis
    Nutrition Reviews · 2019 · meta-analysis · 23 trials

Lavender oil (Silexan)

  1. Lavender oil preparation Silexan is effective in generalized anxiety disorder – a randomized, double-blind comparison to placebo and paroxetine
    International Journal of Neuropsychopharmacology · 2014 · rct · n = 539
  2. Efficacy and safety of lavender essential oil (Silexan) capsules among patients suffering from anxiety disorders: a network meta-analysis
    Scientific Reports · 2019 · meta-analysis · 5 trials · n = 645
  3. Efficacy of orally administered Silexan in patients with anxiety-related restlessness and disturbed sleep – a randomized, placebo-controlled trial
    European Neuropsychopharmacology · 2015 · rct · n = 170
  4. Beneficial effects of Silexan on sleep are mediated by its anxiolytic effect
    Journal of Psychiatric Research · 2019 · rct · n = 212
  5. Silexan in anxiety, depression, and related disorders: pharmacological background and clinical data
    European Archives of Psychiatry and Clinical Neuroscience · 2025 · systematic review · 7 trials

Panax ginseng

  1. Ginseng and ginseng herbal formulas for symptomatic management of fatigue: a systematic review and meta-analysis
    Journal of Integrative and Complementary Medicine · 2023 · meta-analysis · 19 trials
  2. Antifatigue effects of Panax ginseng C.A. Meyer: a randomised, double-blind, placebo-controlled trial
    PLOS ONE · 2013 · rct · n = 90
  3. Efficacy of Korean red ginseng (Panax ginseng) for middle-aged and moderate level of chronic fatigue patients: a randomized, double-blind, placebo-controlled trial
    Complementary Therapies in Medicine · 2020 · rct · n = 50
  4. Effect of Korean Red Ginseng in individuals exposed to high stress levels: a 6-week, double-blind, randomized, placebo-controlled trial
    Journal of Ginseng Research · 2019 · rct · n = 63
  5. The effects of Korean Red Ginseng on stress-related neurotransmitters and gene expression: a randomized, double-blind, placebo-controlled trial
    Journal of Ginseng Research · 2023 · rct · n = 90

Caffeine

  1. Wake up and smell the coffee: caffeine supplementation and exercise performance—an umbrella review of 21 published meta-analyses
    British Journal of Sports Medicine · 2020 · systematic review · 21 trials
  2. The effect of caffeine on subsequent sleep: A systematic review and meta-analysis
    Sleep Medicine Reviews · 2023 · meta-analysis · 24 trials · n = 340
  3. International society of sports nutrition position stand: caffeine and exercise performance
    Journal of the International Society of Sports Nutrition · 2021 · systematic review
  4. Caffeine intake and its influences on heart rate variability recovery in healthy active adults after exercise: A systematic review and meta-analysis
    Nutrition, Metabolism and Cardiovascular Diseases · 2022 · meta-analysis · 12 trials
  5. Caffeine affects autonomic control of heart rate and blood pressure recovery after aerobic exercise in young adults: a crossover study
    Scientific Reports · 2017 · crossover · 1 trials · n = 32

Beta-alanine

  1. β-alanine supplementation to improve exercise capacity and performance: a systematic review and meta-analysis
    British Journal of Sports Medicine · 2017 · meta-analysis · 40 trials · n = 1,461
  2. Effects of β-alanine supplementation on exercise performance: a meta-analysis
    Amino Acids · 2012 · meta-analysis · 15 trials · n = 360
  3. International society of sports nutrition position stand: Beta-Alanine
    Journal of the International Society of Sports Nutrition · 2015 · systematic review

Zinc lozenges

  1. Zinc for prevention and treatment of the common cold
    Cochrane Database of Systematic Reviews · 2024 · systematic review · 34 trials · n = 8,526

Vitamin C (regular use)

  1. Vitamin C for preventing and treating the common cold
    Cochrane Database of Systematic Reviews · 2013 · systematic review · 29 trials · n = 11,306

Magnesium

  1. Oral magnesium supplementation for insomnia in older adults: a systematic review & meta-analysis
    BMC Complementary Medicine and Therapies · 2021 · meta-analysis · 3 trials · n = 151
  2. The role of magnesium in sleep health: a systematic review
    Sleep Medicine Reviews (review) · 2022 · systematic review

Phosphatidylserine

  1. Blunting by chronic phosphatidylserine administration of the stress-induced activation of the hypothalamo-pituitary-adrenal axis in healthy men
    European Journal of Clinical Pharmacology · 1992 · crossover · n = 9
  2. The effects of phosphatidylserine on endocrine response to moderate intensity exercise
    Journal of the International Society of Sports Nutrition · 2008 · crossover · n = 10
  3. A soy-based phosphatidylserine/phosphatidic acid complex (PAS) normalizes the stress reactivity of hypothalamus-pituitary-adrenal-axis in chronically stressed male subjects: a randomized, placebo-controlled study
    Lipids in Health and Disease · 2014 · rct · n = 72
  4. Effects of soy lecithin phosphatidic acid and phosphatidylserine complex (PAS) on the endocrine and psychological responses to mental stress
    Stress · 2004 · rct · n = 80

Echinacea

  1. Echinacea for preventing and treating the common cold
    Cochrane Database of Systematic Reviews · 2014 · rct · 24 trials · n = 4,631
  2. Echinacea for the prevention and treatment of upper respiratory tract infections: A systematic review and meta-analysis
    Complementary Therapies in Medicine · 2019 · rct · 29 trials
  3. Adverse reactions associated with echinacea: the Australian experience
    Annals of Allergy, Asthma & Immunology · 2002 · rct · n = 51

L-carnitine

  1. Effects of L-carnitine supplementation on markers of exercise-induced muscle damage in healthy adults: A systematic review and meta-analysis of randomized controlled trials
    Advanced Exercise and Health Science · 2025 · meta-analysis · 14 trials · n = 284
  2. L-Carnitine Tartrate Supplementation for 5 Weeks Improves Exercise Recovery in Men and Women: A Randomized, Double-Blind, Placebo-Controlled Trial
    Nutrients · 2021 · rct · 1 trials · n = 80
  3. L-Carnitine treatment reduces severity of physical and mental fatigue and increases cognitive functions in centenarians: a randomized and controlled clinical trial
    The American Journal of Clinical Nutrition · 2007 · rct · 1 trials · n = 66
  4. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis
    Nature Medicine · 2013 · cohort · 1 trials · n = 2,595

Rhodiola rosea

  1. Rhodiola rosea for physical and mental fatigue: a systematic review
    BMC Complementary and Alternative Medicine · 2012 · systematic review

Citrulline malate

  1. Acute Effect of Citrulline Malate on Repetition Performance During Strength Training: A Systematic Review and Meta-Analysis
    International Journal of Sport Nutrition and Exercise Metabolism · 2021 · meta-analysis · 8 trials · n = 137
  2. Effects of Citrulline Malate Supplementation on Exercise Performance: A Systematic Review and Three-Level Meta-Analysis
    Nutrients · 2026 · meta-analysis · 30 trials · n = 644
  3. A critical review of citrulline malate supplementation and exercise performance
    European Journal of Applied Physiology · 2021 · systematic review

Lemon balm (Melissa officinalis)

  1. The effects of lemon balm (Melissa officinalis L.) on depression and anxiety in clinical trials: a systematic review and meta-analysis
    Phytotherapy Research · 2021 · meta-analysis
  2. The possible 'calming effect' of subchronic supplementation of a standardised phospholipid carrier-based Melissa officinalis L. extract in healthy adults with emotional distress and poor sleep conditions
    Frontiers in Pharmacology · 2023 · rct · n = 100
  3. Effects of herbal combination (Melissa officinalis L. and Nepeta menthoides Boiss. & Buhse) on insomnia severity, anxiety and depression in insomniacs: randomized placebo controlled trial
    Integrative Medicine Research · 2018 · rct · n = 45
  4. Pilot trial of Melissa officinalis L. leaf extract in the treatment of volunteers suffering from mild-to-moderate anxiety disorders and sleep disturbances
    Mediterranean Journal of Nutrition and Metabolism · 2011 · pilot · n = 20

Quercetin

  1. Quercetin supplementation and upper respiratory tract infection: A randomized community clinical trial
    Pharmacological Research · 2010 · rct · 1 trials · n = 1,002
  2. Effect of Flavonoids on Upper Respiratory Tract Infections and Immune Function: A Systematic Review and Meta-Analysis
    Advances in Nutrition · 2016 · meta-analysis · 14 trials · n = 531
  3. Quercetin and endurance exercise capacity: a systematic review and meta-analysis
    Medicine & Science in Sports & Exercise · 2011 · meta-analysis · 11 trials
  4. Quercetin reduces illness but not immune perturbations after intensive exercise
    Medicine & Science in Sports & Exercise · 2007 · rct · 1 trials

Umbral is a wellness product, not a medical device. It does not diagnose, treat, cure or prevent any disease. Supplement information is educational; talk to a clinician before starting anything, especially if you are pregnant, under 18, managing a condition or taking medication. These statements have not been evaluated by the Food and Drug Administration.

Keep reading