Measuring Your Own Stress

Stress is not a blood test. Cortisol is measurable; "stress" is not — every method on the market measures a proxy, and each proxy answers a different question. This page is the honest owner's manual: what the lab methods actually capture, where consumer kits go wrong, and the free signals that may tell you more than any panel.

Cortisol Is Measurable; "Stress" Is Not

Begin with the uncomfortable sentence: no test measures stress. Tests measure cortisol — a real hormone with real chemistry — and then you, or a marketer, interpret what that number says about stress. The interpretation is where the trouble lives, because cortisol answers at least three different questions depending on how you sample it: what is the hormone doing right now (blood), what shape is its daily rhythm (timed saliva), and what has the average of months been (hair). A single un-timed sample answers none of them well. Before buying anything, decide which question you are actually asking. Most people are really asking a fourth one — "am I too stressed?" — and the free proxies below answer that one better.

Saliva: the Curve in Four Samples

Saliva sampling is the research standard for the daily curve, and the protocol exists precisely because the timing is everything. The consensus guidelines for the awakening response specify a sample at waking — before brushing teeth, since brushing can contaminate the sample with blood — and further samples at +30 and +45 minutes, with exact times recorded, plus an evening sample for the descent (Stalder et al., Psychoneuroendocrinology, 2016). What that protocol buys you is the two shapes from the rhythm page: the morning rise and the evening fall.

Two honest caveats. First, interpretation: labs report reference ranges, but there are no diagnostic thresholds for "stress" — the guidelines describe how to collect data, not how to diagnose you with it. Second, logistics: one day's curve is influenced by that day's sleep, food, and events. One well-collected day tells you about that day; two or three days — ideally a workday and a free day, since the awakening response differs between them (Schlotz et al., Psychosomatic Medicine, 2004) — start to tell you about your pattern. Consumer kits handle the collection fine; the weak link is the interpretation, which they leave to you.

Hair: the Months-Long Average

Hair cortisol is the chronometer of the stress lab. Hair grows about a centimeter a month, and cortisol diffuses into the shaft as it forms — so the three centimeters nearest the scalp hold roughly three months of exposure (Stalder & Kirschbaum, Brain, Behavior, and Immunity, 2012). Its virtue is stability: a bad morning or a rushed sample cannot throw it off. Its limitation is the same trait in reverse — it changes too slowly to track an intervention that works in weeks.

The evidence for what hair cortisol can see is real and group-level. Long-term unemployed adults carry higher hair cortisol than employed controls (Dettenborn et al., Psychoneuroendocrinology, 2010); shift workers and people in chronic high-strain jobs show elevations; a systematic review confirmed the association with stress exposure while cataloguing the confounders — hair treatments, washing frequency, laboratory differences (Staufenbiel et al., Psychoneuroendocrinology, 2013). Those confounders are the caution: a single hair value on a single person is suggestive, not conclusive. As a twice-a-year trend line, though, it is the closest thing the market offers to a "chronic load" readout.

Blood, Urine, and the Point-Sample Trap

A single blood draw is the weakest common measurement, for three reasons that stack: cortisol varies dramatically by time of day, the venipuncture itself can raise it, and one point cannot show a rhythm — a high reading at 8 am and a normal reading at 8 pm are the same molecule in different contexts. Blood cortisol has clinical uses — timed tests are part of the workup for suspected adrenal insufficiency or cortisol excess — but as a wellness metric it is a point sample pretending to be a verdict. The honest clinical tools are different animals: a 24-hour urine collection measures total daily cortisol output, and late-night salivary cortisol is a screening test for cortisol excess, both standard in the evaluation of Cushing's syndrome (Nieman et al., J Clin Endocrinol Metab, 2008). Those tests answer a medical question — does this person have a cortisol disorder — not a lifestyle score. If your motivation is symptoms — rapid weight gain, high blood pressure, purple stretch marks, persistent exhaustion — the next step is a clinician, not a wellness panel.

What Each Method Captures

Read the chart as a question-sorter, not a ranking. If you want the daily rhythm, only timed saliva gets you there. If you want the chronic average, hair is the tool. If you want to know whether you are too stressed, the bottom rows — trends from free proxies — are where the practical answer usually lives.

The Free Proxies That May Matter More

The most useful stress measures cost nothing, because they measure the system's outputs rather than one hormone in it:

• 🛌 Sleep quality first. Waking unrefreshed, fragmented nights, or tired-but-wired evenings are rhythm signals in plain language — the Sleep pillar reads them for you.
• ❤️ Resting heart rate, trending. Measured the same way each morning, a resting rate that climbs over weeks is one of the cleaner recovery signals available — tracked in the quarterly audit.
• 📉 Heart-rate variability. A meta-analysis found chronic stress consistently associated with reduced HRV (Kim et al., Psychiatry Investigation, 2018). The sleep-recovery audit owns HRV tracking on this site, and the vagus-nerve page explains the physiology.
• 📝 The Perceived Stress Scale. Ten questions, two minutes, no equipment — and forty years of validation behind it (Cohen, Kamarck & Mermelstein, J Health Soc Behav, 1983). Your own appraisal, scored honestly, predicts outcomes better than most gadgets.
• 🌙 The evening signal. Tired but wired at 11 pm is the flattening marker from the rhythm page — free, daily, and more informative than most single cortisol values.

A Sane Measurement Protocol

Put it together as a staged plan, and the whole endeavor gets cheaper and more honest at each step:

• Stage 1 — two weeks, free. Log three signals daily: morning energy, evening wiredness (0–10), and sleep quality, plus resting heart rate on three mornings a week. Patterns here answer the question for most people.
• Stage 2 — one saliva day, if it changes a decision. The four-sample protocol on a typical workday, repeated on a free day if the first day looks odd. You are measuring the curve — see the rhythm page for how to read it.
• Stage 3 — hair, at most twice a year. Useful as a trend line across seasons of life, not a monthly score; compare against your own history, not anyone else's.
• Exit criteria. Stop when the data stops producing new decisions — or immediately if checking becomes compulsive. Curiosity has an end date; anxiety sets its own schedule.

And the boundary sentence, stated plainly: measuring for curiosity is fine; measuring because of symptoms is medicine. Persistent exhaustion, sleep disturbance, weight or blood-pressure changes, or mood that won't lift — those are clinician questions, and the right tests are ordered differently than the ones sold in a wellness kit.

Questions, Answered Briefly

• 💊 Should I buy the cortisol panel my feed keeps showing me? Most are single point samples — the weakest possible question, expensively asked. If you test, test the curve: four timed saliva samples.
• 📈 What counts as "high" on these tests? There is no diagnostic threshold for stress — labs give reference ranges, not verdicts. Trend against your own baseline, not against the internet.
• ⌚ Is my watch's "stress score" real? Most are repackaged HRV with proprietary weighting — directionally useful, numerically opaque. Track the trend, ignore the marketing precision.
• 🩺 When is testing a doctor's job instead? When symptoms drive the question — unexplained weight change, hypertension, exhaustion, mood changes. That evaluation looks for cortisol disorders, and it belongs with a clinician.

Related Topics