Ashwagandha, as a standardized root extract at the dose used in trials, is the only supplement in this category with repeated randomized evidence of lowering measured cortisol. Everything else either works indirectly, works on how stressed you feel rather than on the hormone, or does not have the data.

The larger issue sits underneath the product question. Most people shopping for a cortisol supplement have never measured their cortisol, and the symptoms driving the purchase — fatigue, poor sleep, weight around the middle, irritability — overlap almost completely with thyroid dysfunction, iron deficiency, sleep apnea, and depression. A test before a purchase changes the decision more often than not.

The Verdict

KSM-66 ashwagandha at 600 mg daily, or Sensoril at 125–250 mg, is the best-supported choice — expect a modest reduction in morning cortisol over about 8 weeks, not a transformation. Magnesium glycinate at night is the sensible foundation, because it works through sleep. Skip multi-ingredient “cortisol manager” blends that hide per-ingredient doses. Ashwagandha is contraindicated in pregnancy, on levothyroxine, and on immunosuppressants.

The six options that survive scrutiny

Doses below are the doses the underlying trials used, not the doses printed on the average bottle. Ordered by the strength of the direct cortisol evidence.

Supplement and formTrial doseWhat the evidence showsPositioning
Ashwagandha — KSM-66 extract 600 mg once daily, or 300 mg twice daily Root-only extract standardized to ≥5% withanolides. Multiple RCTs show morning serum cortisol falling roughly 15–30% over 8 weeks in chronically stressed adults. Strongest evidence of any option here
Ashwagandha — Sensoril extract 125–250 mg/day Root-and-leaf extract standardized to ≥10% withanolides. More sedating at equivalent dose; trials also report cortisol reduction. Alternative to KSM-66, better suited to evening dosing
L-theanine 200–400 mg/day Reliably lowers subjective stress and raises alpha-wave activity. Effects on measured cortisol are small and inconsistent. Best for acute stress and for smoothing caffeine
Magnesium glycinate 200–400 mg elemental, evening No direct cortisol-lowering trials. Works indirectly by improving sleep, and short sleep is one of the most reliable ways to raise next-day cortisol. The foundation most people skip
Phosphatidylserine 400–800 mg/day Blunts the cortisol response to acute exercise stress in trained subjects at 600–800 mg. Older studies used bovine-derived material; current products are soy- or sunflower-derived and less well studied. Narrow use case: heavy training blocks
Rhodiola rosea (SHR-5 or 3% rosavins / 1% salidroside) 200–400 mg/day, morning Evidence supports reduced fatigue and burnout symptoms. Direct cortisol data is weaker than for ashwagandha. Choose when fatigue, not stress, is the complaint

The two ashwagandha extracts are not interchangeable. KSM-66 uses root only, standardized to at least 5% withanolides, and is the extract in most of the cortisol trials at 600 mg. Sensoril uses root and leaf, standardized to at least 10% withanolides, and is dosed far lower at 125–250 mg because the withanolide concentration is higher. A bottle that says only “ashwagandha 500 mg” with no extract name and no withanolide percentage tells you nothing about whether it matches either.

Safety, interactions, and upper limits

  • Ashwagandha and liver injury. There are documented case reports of drug-induced liver injury, typically presenting as jaundice weeks to months into use. It is rare, but it means people with existing liver disease should avoid it, and new jaundice, dark urine, or right-sided abdominal pain on ashwagandha warrants stopping and getting liver enzymes checked.
  • Ashwagandha and immunosuppressants. It stimulates immune activity, which conflicts directly with tacrolimus, cyclosporine, mycophenolate, and similar drugs after transplant. Avoid.
  • Ashwagandha and sedatives. Additive with benzodiazepines, zolpidem, and alcohol. Expect more sedation than either alone.
  • Ashwagandha in pregnancy. Contraindicated. Traditional use includes abortifacient properties and safety data is absent.
  • Rhodiola. Can cause agitation and insomnia if taken late in the day; several sources note potential interference with CYP-metabolized drugs, so review it against any narrow-therapeutic-index medication.
  • Phosphatidylserine. Doses above 600 mg are associated with GI upset. Most current products are soy-derived, which matters if soy is an allergen for you.
  • Magnesium. Keep total elemental magnesium from supplements under roughly 500 mg per day. It should be separated by 2–4 hours from levothyroxine, tetracycline and quinolone antibiotics, and bisphosphonates, all of which it binds and blocks.

How to test cortisol properly

Cortisol follows a steep daily curve, peaking 30–45 minutes after waking and falling to a nadir near midnight. A single random draw at 2 PM is uninterpretable. These are the tests that mean something. A morning serum cortisol sits inside the base hormone panel of several consumer platforms, so obtaining one does not require a specialist referral: see our platform comparisons for which base panels include it, and book the draw before 9am for the same diurnal reason. The urinary and suppression tests below are a different matter and are physician-ordered.

TestHow it is collectedWhat normal looks likeWhen it is used
Morning serum cortisol Drawn 7–9 AM Typical range roughly 6–23 mcg/dL The standard first test. A single value has wide day-to-day variation
Late-night salivary cortisol 11 PM–midnight, two separate nights Should be at the daily nadir The most useful screen for genuine cortisol excess
4-point salivary cortisol curve Waking, +30 min, afternoon, bedtime Shows the shape of the curve, not just one point Useful for pattern; not a diagnostic test for Cushing syndrome
24-hour urinary free cortisol Full-day collection Integrates total daily output Ordered when excess is suspected clinically
1 mg overnight dexamethasone suppression test 1 mg at 11 PM, serum cortisol at 8 AM Failure to suppress below ~1.8 mcg/dL is abnormal The confirmatory test for Cushing syndrome — physician-ordered
DHEA-S, alongside cortisol Morning draw Interpreted as a ratio with cortisol Adds context; falls with age independent of stress

Two clarifications worth stating plainly. Genuine cortisol excess — Cushing syndrome — is a diagnosable medical condition with a defined workup, and it is not treated with supplements. The point at which this stops being a supplement question is a late-night salivary cortisol that is not at its nadir across two separate nights, since that is the screen for genuine excess; confirmation runs through a 24-hour urinary free cortisol or the 1 mg overnight dexamethasone suppression test, both of which a physician orders and interprets. At the other end, “adrenal fatigue” is not a recognized diagnosis; systematic reviews have found no consistent evidence that the pattern exists as described. What does exist is a real, measurable elevation of cortisol from chronic short sleep, alcohol, overtraining, and untreated sleep apnea, and that is worth finding because it has real treatments.

What to buy, and what to check on the label

  1. A named, standardized extract. KSM-66 or Sensoril for ashwagandha; SHR-5 or a stated 3% rosavins / 1% salidroside for rhodiola. Generic “ashwagandha powder” is not the material the trials used.
  2. The withanolide percentage. If it is not printed, the potency is unknown. 5% for a root extract, 10% for root-and-leaf.
  3. A per-ingredient dose, not a blend weight. The word “proprietary blend” on a cortisol product is the single most reliable signal to put it back.
  4. Third-party verification. USP, NSF Certified for Sport, or Informed Sport. Botanical extracts are a common site of adulteration, so this matters more here than in most categories.
  5. Elemental content on any magnesium. A 400 mg magnesium glycinate capsule contains roughly 50–60 mg of elemental magnesium. See the magnesium guide for the full breakdown.

What lowers cortisol more than any of this

Ranked by effect size, the non-supplement interventions dominate. Seven or more hours of sleep, since even one night at four hours measurably raises next-day cortisol. No caffeine after early afternoon, given a 5–6 hour half-life. Less alcohol, which fragments the second half of the night. Regular Zone 2 aerobic work, which lowers the cortisol response to a given stressor over 8–12 weeks. And treating sleep apnea if you have it — untreated apnea raises overnight cortisol and no supplement compensates.

Frequently Asked Questions

What is the best supplement for lowering cortisol?

Ashwagandha, specifically a standardized root extract such as KSM-66 at 600 mg daily or Sensoril at 125–250 mg daily. It is the only option in this category with several randomized trials showing measured reductions in morning serum cortisol, generally in the range of 15–30% over 8 weeks in adults reporting chronic stress. Everything else here is second-line, and none of it approaches the effect of fixing sleep.

How long does ashwagandha take to lower cortisol?

Trials typically measure at 8 weeks, and most report the change at that point. Subjective stress and sleep quality often improve within 2–3 weeks. If nothing has changed at 8–12 weeks on a full labeled dose of a standardized extract, the supplement is not the answer for you and continuing costs money without benefit.

Should I test my cortisol before supplementing?

It changes the decision, so yes. A morning serum cortisol establishes whether your level is even elevated, and a late-night salivary cortisol screens for genuine excess. Most people who buy cortisol supplements have normal cortisol and a sleep, caffeine, alcohol, thyroid, or iron problem instead. Direct-to-consumer panels and physician-led programs both include cortisol; the useful part is having a number before and after.

Who should avoid ashwagandha?

Several groups. It is contraindicated in pregnancy. It should be avoided by anyone on levothyroxine or with hyperthyroidism, because it raises T3 and T4 and has produced thyrotoxicosis. Anyone on an immunosuppressant after transplant should avoid it, since it stimulates immune activity. It adds to the effect of sedatives and benzodiazepines. It also has documented case reports of liver injury, so anyone with existing liver disease should not take it without physician oversight.

Do cortisol blocker or cortisol manager blends work?

Usually not, for a structural reason. These products list eight to twelve ingredients and a total blend weight rather than per-ingredient amounts. That format exists so a manufacturer can include a trial-backed ingredient at a fraction of the trial dose. If the label does not show a named, standardized ashwagandha extract at 300 mg or more as its own line item, assume the dose is below what the studies used.

What lowers cortisol more than any supplement?

Sleep, by a wide margin. A single night restricted to four hours raises next-evening cortisol measurably, and chronic short sleep keeps it elevated. After that: stopping caffeine by early afternoon, since caffeine has a half-life of roughly 5–6 hours and raises cortisol directly; reducing alcohol, which fragments sleep architecture; and regular aerobic training, which lowers the cortisol response to a given stressor over 8–12 weeks. Any one of those outperforms every product on this page.

How long can I keep taking ashwagandha, and when should I stop?

The trials behind the cortisol effect generally measure at 8 weeks, so continuous use well beyond that sits outside what the evidence describes rather than being something shown to be unsafe. Two stop rules follow from the rest of this page. The first is the 8–12 week no-effect rule set out above, which is a reason to stop rather than to persist. The second is that because the reported cases of liver injury typically appear weeks to months into use, new jaundice, dark urine, or right-sided abdominal pain is a reason to stop and have liver enzymes checked rather than to wait it out. There is no established cycling schedule, so anyone intending to take it indefinitely should review it with their prescriber as they would a medication.

Related