Cortisol gets discussed as though it were a toxin: something to lower, flush, detox, or block. That framing gets the biology backwards. Cortisol is one of the body’s primary timing and mobilization signals, it follows a steep and predictable daily rhythm, and a morning without a cortisol rise would be a morning you could not get out of bed for.
The honest version of the sleep problem is not “my cortisol is too high.” It’s a question of when. Cortisol is supposed to peak shortly after you wake and reach its lowest point in the first hours of sleep. Sleep onset happens at the bottom of that curve. When falling asleep is difficult in the wired-but-tired way, what has usually happened is that the evening end of the curve is still elevated — not because a gland is broken, but because the day kept feeding it.
This page is the hormone underneath several others in the wind-down cluster. Circadian rhythm explains the clock that sets the curve’s timing. This one explains what the curve is made of, and which parts of it you can actually reach.
What cortisol actually is
Cortisol is a glucocorticoid hormone, produced by the outer layer of the adrenal glands, which sit on top of the kidneys. Its release is controlled by a three-part chain usually called the HPA axis: the hypothalamus signals the pituitary, the pituitary releases ACTH into the bloodstream, and ACTH tells the adrenal cortex to produce cortisol. Cortisol then feeds back on the hypothalamus and pituitary to slow the chain down. It is a thermostat, not a tap.
What cortisol does once it’s circulating is mostly about making energy and attention available. It raises available blood glucose, supports blood pressure and cardiac output, modulates immune and inflammatory activity, and sharpens vigilance. Under acute stress that combination is exactly what you want, which is how it acquired the “stress hormone” nickname. But the same combination is what gets you through a normal Tuesday, and most of the cortisol you produce has nothing to do with stress at all.
Two things follow from that. The first is that “high cortisol” is meaningless without a time of day attached. The second is that if you’re trying to be less activated in the evening, you’re working on a system whose whole purpose is mobilization. It doesn’t turn off. It descends.
The daily curve
The shape of the curve is one of the more reliable findings in human physiology. Cortisol is at its lowest in the first hours of sleep, begins rising in the second half of the night, climbs steeply in the 30 to 45 minutes after you wake, then falls across the day with a long tail into the evening.
That post-waking surge has its own name: the cortisol awakening response. Pruessner and colleagues described it in the late 1990s, and it’s been studied heavily since, partly because it’s convenient to measure and partly because it seems to be sensitive to psychological state. It is triggered by waking itself, not by getting up or by light exposure, though light reinforces it. A morning that feels heavy and unstartable is often a morning where this response is blunted, which is one of the more concrete arguments for a stable wake time: the surge is anchored to when you wake, so an erratic wake time produces an erratic morning.
The evening end matters just as much and gets far less attention. Across the afternoon cortisol declines, and by late evening it should be near its floor. That descent overlaps with the rise in melatonin and the drop in core body temperature, which is why the hour before bed feels physiologically distinct rather than just conventionally late. The transition through the hypnagogic state sits at the bottom of all three curves at once.
Which reframes what a wind-down routine is doing. It isn’t a mood. It’s the behavioral protection of a window that already exists.
Wired but tired
“Wired but tired” is a precise description of a real state, and it happens because sleepiness and alertness are produced by different systems rather than by one dial.
Sleep pressure builds through the day as adenosine accumulates in the brain — the longer you’ve been awake, the heavier it gets. Arousal is separate: sympathetic nervous system activity, circadian alerting signal, and cortisol among the inputs. You can have both high at once, and when you do, the pressure loses. You feel exhausted, your eyes hurt, and you’re wide awake, because a mobilized body doesn’t hand over to sleep just because it’s tired.
This is also why the last hour of the evening decides so much. If you spend it working, arguing, scrolling something aggravating, or planning tomorrow in detail, you’re supplying fresh input to the arousal side at the exact moment the other side needs an uncontested run. Overthinking at night is the version of this that happens after lights-out, and it’s effective at keeping people awake for the same reason: rumination is activation, and it keeps the handover from completing.
The second wind fits here too. Late in the evening the circadian system produces an alerting signal before the drop into sleep, and if you push through your window you arrive on the far side of it feeling suddenly capable. Nothing has improved. You’ve just stopped being offered the descent for a while.
Cortisol and insomnia: a loop, not a cause
The research on cortisol and sleep is genuinely bidirectional, which is worth stating plainly because both halves get sold separately.
Vgontzas and colleagues reported in 2001 that people with chronic insomnia showed elevated HPA-axis activity across the night compared with good sleepers — a pattern consistent with insomnia as a disorder of hyperarousal rather than of insufficient sleepiness. That framing has held up well and it’s the reason the effective treatments for insomnia are arousal-reducing rather than sedating.
Running the other way, Spiegel, Leproult and Van Cauter found in 1999 that restricting healthy young men to short nights slowed the evening decline in cortisol, alongside changes in glucose handling. Short sleep changed the endocrine profile within days. Sleep loss isn’t only a consequence of an activated system; it’s an input to one.
Put together, this is a loop rather than a first cause. A stressful stretch raises evening arousal, which delays sleep onset, which shortens sleep, which flattens the next evening’s descent, which delays onset further. It’s also why the loop can be entered from either side and, more usefully, interrupted from either side. Protecting sleep duration is a way of lowering evening cortisol, not just a beneficiary of it.
What’s oversold
The gap between this physiology and the content built on it is wide, and it’s worth sorting.
“Adrenal fatigue” is the biggest claim and the weakest. The proposal is that prolonged stress exhausts the adrenal glands until they can’t produce enough cortisol, causing fatigue, brain fog, and poor sleep. A 2016 systematic review by Cadegiani and Kater examined the literature and found no evidence supporting it as a condition, and the Endocrine Society does not recognize the diagnosis. Adrenal insufficiency is real — Addison’s disease is a serious, treatable illness — but it is diagnosed by specific testing, not by a symptom questionnaire on a supplement site. The reason the label persists is that the symptoms it describes are extremely common. Being exhausted and foggy is real. The mechanism offered for it isn’t.
Home cortisol panels are difficult to use well. Because the curve is so steep, especially in the first hour after waking, sampling half an hour off schedule moves the number more than most of the things you’d want to detect. A single reading tells you very little; a properly interpreted diurnal profile requires timing discipline and a clinician who knows what to do with the result.
Cortisol-lowering supplements are a mixed picture. Ashwagandha has the most human trial evidence in this space, and several small randomized trials have reported reductions in stress measures and in cortisol. The trials are small, short, often industry-funded, and heterogeneous in preparation and dose, so the honest summary is “plausible and modest” rather than established. Everything else in the category is weaker. As with magnesium and melatonin, the supplement question is worth asking only after the behavioral levers are actually in place, because it’s the cheapest question to ask and the least likely to be the answer.
“Detoxing” cortisol isn’t a thing. Cortisol is metabolized by the liver and cleared continuously. There is no accumulation to flush.
The wind-down window is a timing problem
If nothing lowers cortisol on command, the practical question becomes which inputs shift the curve, and when they have to happen. Most of them are earlier in the day than people expect.
Wake time anchors the whole curve. The awakening response is keyed to when you wake, and the rest of the day’s slope descends from it. A consistent wake time — including weekends, within an hour — does more for the evening end than anything you can do in the evening itself. Morning light shortly after waking reinforces it.
Caffeine raises cortisol, and its half-life means an afternoon coffee is still meaningfully present at bedtime. It also blocks adenosine, so it works against sleep pressure and for arousal at the same time. Early afternoon is a reasonable cutoff for most people.
Hard exercise acutely raises cortisol, which is a normal, healthy response and part of why training works. The timing note is narrow: intense sessions late in the evening leave activation running into the window you need for the descent. Morning or early evening avoids the conflict, and gentle movement at night is fine.
Alcohol is the common miss. It shortens sleep onset and then fragments the second half of the night, and it raises cortisol as it clears. The sedation is real and the cost arrives at 3am.
The last hour is the lever you control most directly. Slow, exhale-led breathing engages the parasympathetic side through the vagus nerve, and it works within minutes: 4-7-8 breathing or anything in the breathing exercises for sleep family will do. This won’t reshape the day’s whole curve. What it does is stop adding to the arousal side at the point where that matters most.
What to do this week
Pick the timing fix, not the hormone fix.
For one week, hold your wake time within a thirty-minute band and get outside, or at a bright window, within half an hour of getting up. That’s the single intervention with the widest downstream effect, because it anchors both ends of the curve. Nothing else on the list works as reliably.
Then protect a real window before bed. Thirty minutes, phone out of reach, nothing that requires a decision. Two minutes of breathing in for four and out for six at the start of it — soft, not effortful, since strain is itself activation. If you’re the kind of person whose mind fills that space with tomorrow, write the list down before the window opens rather than trying to suppress it during; that’s the practical overlap with worry and the reason a bedtime routine is sequenced rather than just relaxing.
One thing to stop doing: don’t try to assess whether your cortisol is coming down. Monitoring arousal is arousal, and it’s the same trap that sleep anxiety runs on. The evidence you want is how the next few mornings feel, not a reading.
And if a period of poor sleep has stretched past a few weeks and is affecting your days, that’s a clinician’s conversation about insomnia rather than a hormone panel. CBT-I is the first-line treatment, and it works on exactly the hyperarousal this page is about.
How Murmora fits
The part of the night Murmora is built for is the bottom of the curve — the short window where arousal has already started dropping and the mind is unusually receptive but hasn’t yet been given anything to do. That’s the window most people spend on tomorrow’s list. Murmora fills it with a sparse whisper instead: a few sentences at the threshold, paced for a body that’s settling, written from what you actually said you’re working on rather than from a generic script.
You start with one of a small set of guide voices chosen for unhurried pacing, and the same session can later be regenerated in your own cloned voice. The timing work in this article is what opens the window. A quiet sleep affirmation is what you put in it.