Bought a wellness gadget on Black Friday? How to tell if it's actually working for your body

Short answer

You find out by testing it on yourself, not by trusting the average. Pick one or two things you can measure at home — sleep, resting heart rate, heart rate variability (HRV), a symptom you rate each day — and record them for one to two weeks before you start. Then use the new thing consistently for a set period, keeping the rest of your life as steady as you can, and compare the two stretches using weekly averages, not single days. If the change is bigger than your normal day-to-day wobble, and it holds when you test again, it is probably working for you.

Why bother? Because the study on the box describes an average across many people, and the same supplement, routine or device can do a lot for one person and nothing for the next (1, 2). It also protects you from three honest traps: placebo, random noise and regression to the mean — the tendency of an unusually bad week to be followed by a more ordinary one, whatever you bought (4).

If the thing you bought "should" work and you can't feel it, that is not your fault and it is not a lack of willpower. Bodies differ, timelines differ, and feelings are a noisy instrument. Before you blame yourself, check what your body already recorded.

See Welltory Premium with the blog reader discount

The annual plan renews automatically until you cancel; the lifetime plan is a one-time payment.

Note: this guide is for general education and is not medical advice. If you are trying something to deal with a symptom, a diagnosis or a medication, talk to your clinician before you start and before you stop anything you were prescribed. This guide is published by Welltory, which makes one of the apps mentioned. Welltory is a general wellness product: it helps you track and understand your data, and it does not diagnose, treat or replace medical care.

Why does the same supplement or gadget work for one person and not another?

Most wellness claims come from studies that report one number: the average change in the group that got the thing compared with the group that did not. An average can hide a wide spread. Some people in the treated group improved a lot, some a little, some not at all, and a few got worse.

Some of that spread is real. In a well-known Israeli study, 800 people wore continuous glucose monitors for a week while researchers logged 46,898 meals. They "found high variability in the response to identical meals": the same standardized meal pushed blood sugar up sharply in some people and barely at all in others (2). The same logic applies to wellness purchases. People differ in:

• Where they start. A supplement that corrects a shortfall can make a visible difference for someone who is low and very little for someone who is not. That is why the timelines and tests in our guide to how long supplements take to work start with "do you actually need it?" — and why our guide on whether you should take magnesium begins with who is likely to benefit.

• What else is going on. Sleep debt, a stressful month, illness, alcohol, a training block or the menstrual cycle can swamp the effect of anything you add.

• How they use it. Dose, timing and consistency. A meditation app opened twice in a month is a different intervention from one used every evening.

• Body and context. Age, medications, health conditions and genetics all shape the response. Advice built for an imagined average user often misses these, a problem we wrote about in why your health app's advice doesn't fit you.

Some of that spread is not real. This is the part most marketing skips. Atkinson and Batterham showed that a lot of what looks like "responders" and "non-responders" in a study can be produced by random day-to-day variation and measurement error alone. If you measure anyone twice, the numbers differ even when nothing has changed, so a study will always appear to contain people who "responded" and people who "didn't" (1). The same applies to you: one good morning after a new habit is not proof, and one bad morning is not a verdict.

That is the whole case for testing on yourself. The study tells you whether something can work on average. Only your own, carefully collected data can tell you whether it is working for you — and even then, only if you collect it in a way that separates a real change from ordinary noise.

What changes in your body can you actually measure at home?

More than people think, but not everything that a product promises. A useful rule: if you can't measure it, you can't test it — you can only hope. Before you start, translate the promise on the box into something you can record.

Pick one main outcome and, at most, one or two supporting ones. The main outcome is the thing you bought the product for. If you bought a sleep gadget, it is sleep — not your step count. If you track ten numbers, one of them will move by chance and look like a result.

Decide how you will measure it, and keep it identical. For HRV, that means the same time each morning, before coffee, exercise or a shower, in the same position. Welltory's measurement guide suggests staying in bed for 5–10 minutes before a morning measurement (16). If you measure with your phone camera, read how that compares with a wearable in our guide to how accurate HRV from a phone camera is. A device you already own is usually enough: you do not need a new device to run a good test — see do you need a new wearable.

Don't ignore how you feel — just write it down. Subjective ratings are legitimate outcomes; for sleep, mood or pain they are often the main outcome. They are also the outcomes most sensitive to expectation, which is why it helps to pair them with at least one number your expectations cannot easily move, like resting heart rate. When the two disagree, it is worth knowing why — we explain the common reasons in why your wearable and your feelings disagree.

How long should you test something before deciding it works?

Long enough for the thing to act and long enough for you to see past the noise. Those are two different clocks.

Clock 1: how fast the thing can work. This depends on what you bought.

• Things with an immediate effect — a breathing session, a sauna visit, a nap, a cup of coffee. You can see a short-term change within minutes to hours. For sauna, how long to stay in and what happens afterwards is covered in how long to stay in a sauna.

• Habits — meditation, a new bedtime routine, a fitness class. Benefits build over weeks. Meditation trials typically run around eight weeks, and that is where the clearest effects on anxiety and depression were measured (9).

• Supplements — from a few weeks to several months, depending on the nutrient and on whether you were low to begin with. Rather than repeat it here, use the supplement-by-supplement timelines in our supplement timeline guide.

• Gadgets that measure rather than act — a ring, a watch, a cuff. These don't change your body by themselves. What you are testing is whether having the data changes what you do, and whether that change helps. Give it at least a month of real use.

Clock 2: how long you need to see past the noise. Day-to-day numbers jump around. In elite endurance athletes, researchers found that rolling weekly averages of morning HRV tracked training changes more reliably than single-day values (7), and a follow-up study concluded that at least three valid measurements per week were needed for a usable weekly average (8). Those were athletes, but the principle travels: compare weeks with weeks. We explain why single readings swing so much in why HRV fluctuates day to day.

A practical minimum: one to two weeks of baseline, then at least as long on the new thing as it plausibly needs to work — and never less than two weeks — with at least three measurements a week. Write your end date in your calendar on day one. Deciding in advance when you will judge it stops you from quitting after a bad Tuesday or keeping something for a year because "it might be starting to work".

How to run a simple before/after experiment on yourself

Researchers have a name for a trial with one participant: an n-of-1 trial. A user's guide to these trials, published by the US Agency for Healthcare Research and Quality, says that "systematic assessment of outcomes may well be the single most important design element" and that "repetition is to n-of-1 trials what sample size is to parallel group RCTs" (3). You don't need a lab to borrow those two ideas.

Step 1. Write down the question and the rule. One sentence: "Does 10 minutes of meditation in the evening help me fall asleep faster?" Then the rule you will use to judge it, before you have any data: "I'll call it working if my weekly average time to fall asleep drops by at least 10 minutes and my morning rested rating goes up by at least one point." Setting the bar in advance is your best protection against seeing what you hoped to see.

Step 2. Baseline week (or two). Change nothing yet. Record your main outcome and your supporting numbers every day, the same way, at the same time. Tag or note anything unusual: a late dinner, alcohol, a hard workout, a bad night with a child, illness, travel. At the end, look at the range: your best and worst days, and the weekly average. That range is your personal noise level.

Step 3. Start the change — and only that change. If you start magnesium, a new bedtime and a new training plan in the same week, you won't know what helped. Keep caffeine, alcohol, exercise and bedtimes as close to your baseline as real life allows. Keep using it as intended: an experiment on something you used half the time tests the half.

Step 4. Keep measuring exactly as before. Same time, same position, same device, same rating scale. If you switch from a phone camera to a new ring halfway through, you are comparing two instruments, not two periods.

Step 5. Compare weeks, not days. Put your baseline weekly average next to the averages from the later weeks of the test. The first few days are often the least representative — novelty and expectation are strongest then. Ask three questions:

1. Did the main outcome move in the direction you hoped?

2. Is the change bigger than your baseline wobble? If your baseline weeks already varied by five minutes to fall asleep, a three-minute improvement is inside the noise.

3. Did it hold for more than one week?

Step 6. Check it by repeating. This is the step that turns a hunch into evidence. Stop the change for a week, if it is safe and sensible to do so, and see whether the outcome drifts back. Then restart. If the numbers improve each time you use it and slip each time you stop, that is far stronger than a single before/after. The n-of-1 guide calls this alternating pattern a crossover, and recommends repeating it (3). Don't do this with anything a clinician prescribed: stopping a medication is a decision for them, not for an experiment.

Step 7. Decide by your rule. Keep, adjust or stop — based on what you wrote in step 1, not on how you feel the morning you decide.

What does a real test look like? Meditation, step by step

Here is the whole method applied to one of the most common holiday purchases: an annual subscription to a meditation app (a meditation headband would work the same way). The numbers below are a hypothetical illustration of how to read a result, not real data.

The question. Maya, 38, sleeps badly when work is busy. She buys a meditation app in November and wants to know: "Does a 10-minute guided meditation before bed help me fall asleep faster and wake up more rested?"

Why meditation is a good — and tricky — test case. There is reasonable evidence that it can help some people. A review of 47 trials with 3,515 participants found moderate evidence that mindfulness programs improved anxiety and depression at eight weeks, with small to moderate effects, and no evidence that they were better than active treatments such as exercise or therapy (9). For sleep, a meta-analysis of 18 trials with 1,654 participants found mindfulness meditation improved sleep quality compared with time- and attention-matched controls, but not compared with evidence-based treatments for sleep problems (10). And on HRV, a meta-analysis of 19 randomized trials found "insufficient evidence" that mindfulness and meditation programs raise vagally mediated HRV compared with control conditions, with results that varied widely between studies (11). So it might help Maya's sleep; it may or may not move her HRV; and only her data can say.

Her rule, written on day one. Main outcome: minutes to fall asleep (her estimate, noted each morning) plus a 1–5 "how rested" rating. Supporting: morning HRV and resting heart rate, measured in bed with the same device. "If, in weeks 3–4, I fall asleep at least 10 minutes faster on average and my rested rating is up by at least one point compared with baseline, the app stays. If only HRV moves, that's interesting but not the reason I bought it."

Baseline: two weeks. No meditation. She falls asleep in 25–50 minutes, averaging about 38. Her rested rating averages 2.4. She tags two late work nights and one evening with wine.

The change: four weeks. Ten minutes of guided meditation in bed, every evening, same time. Bedtime, caffeine and exercise stay roughly the same. She also uses Welltory's Experiments a few times to see the short-term effect: an HRV measurement before the session, a tag, and a second measurement afterwards, which the app compares in a report (13). She treats each one as a snapshot of one evening, and looks for whether the direction is consistent over several sessions rather than at any single result.

The comparison. Week 1 looks great — 24 minutes on average — but she knows the first week is when novelty and expectation are strongest. Weeks 3–4 average about 27 minutes, with a rested rating of 3.3. That is 11 minutes faster, outside her baseline range of weekly averages, and it held for two weeks. Her morning HRV weekly average barely changed.

The check. Over the holidays she pauses for a week. Falling asleep drifts back towards 35 minutes. When she restarts, it comes down again.

Her verdict. By her own rule, the app is working for her sleep — even though HRV, the number she assumed would prove it, didn't move. That is a perfectly good result. Had weeks 3–4 looked like baseline, the honest verdict would have been "not for me, in this form", and the next step would be to change one thing (a different time, a different style of practice) or move on.

What this example does not prove. Maya can't rule out that believing in the app helped. For her, that hardly matters: she wanted to fall asleep faster, and she does. What she can rule out is that she was fooling herself with one lucky week.

Black Friday for Welltory users: a special discount on Premium

From November 25 to December 1, 2026, use the code WelltoryBF40 at checkout. You see the final price before you pay. Premium is billed annually and renews automatically until you cancel.

Get the Black Friday offer

Why can't you trust a simple before and after?

Because three things can make an ineffective product look like it works — and none of them means you are gullible.

Placebo and expectation. Expecting to feel better can make you feel better, and it can change how you rate your day. A Cochrane review of trials comparing placebo with no treatment found that placebo effects were modest and showed up mainly in outcomes people report themselves, especially pain and nausea, rather than in outcomes an observer measures (5). The effect can be real even when people know: in a three-week trial in 80 people with irritable bowel syndrome, 59% of those knowingly taking placebo pills reported adequate relief, compared with 35% of those with no treatment (6). For you, this means two things. Subjective ratings deserve a partner number, such as resting heart rate, that expectation moves less easily. And a benefit that is partly expectation is still a benefit — it just may not be worth premium money if a cheaper ritual gives you the same thing.

Noise. Your body is never the same two days running. Sleep, heart rate and HRV move with stress, meals, alcohol, illness, temperature and the menstrual cycle, and devices add their own measurement error. That is why the baseline week matters: it tells you how big a change has to be before it is more than ordinary wobble (1).

Regression to the mean. This is the sneakiest one. People tend to buy a fix when things are at their worst — the week of terrible sleep, the month of constant headaches. But "unusually large or small measurements tend to be followed by measurements that are closer to the mean", and researchers warn that this "can make natural variation in repeated data look like real change" (4). In plain words: a bad stretch is usually followed by a less bad one, with or without the gadget. If you started your test during your worst week, part of any improvement would have happened anyway. A baseline of one to two ordinary weeks, rather than starting on the worst day, is the cheapest protection against this.

Two smaller traps. Measurement reactivity: simply paying attention — logging bedtimes, measuring every morning — can change behaviour, so some of the improvement may come from the tracking itself. Season and life: testing a sleep product across the December holidays, or an energy supplement as days get longer, mixes the product with the calendar. Note these honestly; the repeat step is your best defence.

What if the data says it isn't working?

First, it is not a personal failure. You ran a better test than most product claims ever get. Then work through the list before you decide.

Was the test fair? Did you actually use it as intended, most days? Did something big change during the test — illness, a new job, a holiday, a new medication? Did you give it the time it plausibly needs? For supplements, check the timeline in the supplement timeline guide before you judge.

Were you measuring the right thing? A sleep gadget may not change your HRV at all and still help you fall asleep. A tracker may not improve any number but may help you notice a pattern you can act on. Go back to the reason you bought it.

Was the change hidden in the noise? If your baseline varied a lot, a modest effect is hard to see in four weeks. You can extend the test, or repeat the on/off pattern once more.

Then decide — calmly.

• No change, and it costs money every month: stopping is a reasonable, grown-up decision. Check the seller's return or refund policy and cancel any subscription before it renews; most wellness subscriptions renew automatically.

• No measurable change, but you enjoy it: that is a valid reason to keep something. Just keep it for the right reason, and know what you are paying for.

• A small change you can live without: compare the cost with cheaper ways to get the same effect. Our guide to health investments under $100 and our comparison of stress and sleep apps vs gadgets vs supplements can help, as can the full picture of the real cost of health tracking.

• You feel worse: stop, and if it was a supplement, tell your pharmacist or clinician what you took. Meditation is usually low-risk, but NCCIH notes that in one review about 8% of participants reported a negative effect, most often anxiety or depression (12).

Don't let the experiment become the problem. If checking your numbers starts to make you anxious, or you feel you are failing your own data, take a break. We wrote about this in does health tracking cause anxiety. The point of the test is a decision, not a lifelong audit.

Living with a new wellness purchase: the questions people ask next

Can I test two things at the same time?

Not if you want to know which one helped. If you start a supplement and a meditation app in the same week and your sleep improves, you have no way to separate them. Test one, decide, then add the next. If you must start two at once — for example, a clinician asked you to — at least stagger them by a couple of weeks so each has its own "before".

Should I test a supplement this way or just get a blood test?

Both have a place. For nutrients like iron, vitamin D and B12, a blood test before and after is the most direct way to know whether your levels changed, and your clinician can tell you whether testing makes sense. A home experiment answers a different question: does the change in your levels, or the supplement, make a difference you can feel or measure in daily life? Timelines for each supplement are in our supplement timeline guide.

My new wearable says my sleep is worse than I thought. Is it working?

A wearable that measures does not change your body by itself, so "working" means: is it accurate enough to be useful, and does it help you act? Compare its numbers with how you feel over a couple of weeks, and look at trends rather than single nights. If a scary nightly number keeps you awake, that is a cost, not a feature.

Is a gym or sauna membership worth testing the same way?

Yes, and it often works well because the "dose" is easy to count: visits per week. Pick an outcome you care about — resting heart rate, sleep, a repeatable fitness test, how you feel after work — record a baseline, then compare weeks with and without regular visits. For sauna-specific questions, such as session length, see how long to stay in a sauna.

What if my HRV goes up right after a session but nothing changes long term?

That is common. A short-term change during or right after a breathing session, meditation or sauna is real, but it is not the same as a lasting shift in your baseline. Use the short-term result to see how your body responds in the moment, and the weekly morning average to see whether anything is building.

Is it a waste of money if the effect is "only" placebo?

Not necessarily. If you feel better and the product is safe, you have gained something. But a known placebo-type effect can also come from cheaper rituals, so it is worth asking whether the price matches what you are getting.

Before you blame yourself

When a purchase doesn't deliver, the easy story is a personal one: "I didn't stick with it", "my body is broken", "everyone else gets results". The data usually tells a kinder and more useful story. Maybe the dose of the habit was too small. Maybe the timeline was too short, or the test ran through the worst month of your year. Maybe the product helps the average person a little and you are simply not the person it helps most — which is information, not failure.

You are not broken. Pick one change, measure it the same way for a few weeks, and judge it by the rule you set in advance. That trend is better evidence about your body than any review, any influencer and any single morning.

When to see a doctor

Talk to your clinician before you start if you are buying something to manage a symptom or condition rather than for general wellbeing — for example, sleep problems that have lasted for months, ongoing fatigue, palpitations, or high blood pressure readings. Check with a doctor or pharmacist before adding a supplement if you take prescription medicines, are pregnant or breastfeeding, or have a chronic condition. Never stop or change a prescribed medication as part of a home experiment.

See a doctor, too, if your experiment turns up something unexpected and persistent: a resting heart rate that stays well above your usual for more than a couple of weeks without a clear reason, a run of high blood pressure readings, or a symptom that is getting worse rather than better. If meditation or any other practice leaves you more anxious or low for more than a few days, stop and talk to someone.

Chest pain or pressure, severe shortness of breath, fainting, or a racing or irregular heartbeat that does not settle are reasons to seek urgent care, not to keep testing.

How to bring this up with your doctor

Say it in one sentence. "I've started (product) to help with (outcome), and I tracked (numbers) for two weeks before and four weeks after — can we look at whether it's helping and whether it's safe for me?"

What to bring. The product name and dose; when you started; your baseline and test-period averages; any side effects; all your medicines and supplements. A CSV export or a few screenshots of weekly averages is far more useful than a single "good" or "bad" day.

What to ask. "Is there a test that would show whether this is doing anything?" "Could it interact with anything I take?" "How long would you expect it to take?" "Is there a better-proven option for what I'm trying to fix?"

Start understanding your body

Learn what affects your energy, stress, sleep, and daily state. Get the app.

Explore your data

How Welltory helps — and what it cannot do

The limits first. Welltory is a general wellness app, not a medical device. It does not diagnose, treat or prevent any condition, it cannot tell you whether a supplement is safe for you, and it does not measure nutrient levels, blood pressure or anything a lab test measures. It helps you track and understand the signals your body already produces, so that you can run the kind of test this guide describes on your own data.

Short-term tests with Experiments. Experiments let you see what a single activity does in the moment. You take an HRV measurement before the activity — a breathing practice, a meditation, a bath, a workout — tag what you did, then take a second measurement within three hours, and Welltory shows a report on what changed. The Help Center advises testing one activity at a time and, after a workout, waiting until your heart rate has come back down before the second measurement (13). Experiments are part of the free trial and then of Premium: according to the Help Center, if you don't upgrade, experiments aren't available after the trial (15). On iPhone with an Apple Watch there is also a Self-Discovery experiment: you pick a 1–3 hour stretch of your day and Welltory walks you through how what you did affected your stress and energy; free users get the first one with the full result, and Premium users can run as many as they like (14).

Long-term tests with your baseline. For the weeks-long comparison, measure HRV each morning the same way (16). On iPhone with an Apple Watch or Oura, the Today screen compares today's HRV with your baseline — the average of your past 60 days (18). Weekly and monthly trends and a history kept beyond 30 days are Premium; on the free version, data older than 30 days is deleted (15) — which matters for a test with a two-week baseline and a four-week trial. What the free version keeps is covered in is Welltory free.

Tag what you tried. On iPhone, when Welltory detects a stress or rest episode it asks "What happened?" or "What helped you rest?", and you can tap, type or say your answer — "meditation", "magnesium", "sauna", "late wine". In My Patterns, Welltory collects those tags and shows which ones tend to come with stress or recovery; insights usually need at least 7 occurrences of a tag in the current month plus some history from the previous one (17). Welltory only knows what you tried if you tag it.

A record for your doctor. From the Welltory web app you can export your data as a CSV file (Dashboard → choose a chart and time range → Export) (19), and bring your baseline and test-period averages to an appointment. Our guide to exporting your health data shows how.

What it cannot do. Welltory cannot blind you, so it cannot remove placebo; it cannot control the rest of your life; and a pattern in your data is a connection worth discussing, not proof that the product caused the change. Phone-camera measurements are spot readings, not 24/7 tracking. You do not need a new device to start — your phone camera, or a watch you already own, is enough for the morning routine.

Premium annual is $99 a year on Welltory's plans page (checked October 22, 2026; billed annually, auto-renews unless cancelled); prices in the App Store and Google Play may differ, and a lifetime option is described on the plans page.

Try Premium with the blog reader discount

The annual plan renews automatically until you cancel; the lifetime plan is a one-time payment.

A note for Black Friday: from November 25 to December 1, 2026, Welltory users can use the code WelltoryBF40 at checkout for a special Black Friday discount on Welltory Premium. Premium is billed annually and renews automatically until you cancel, and the checkout page shows the final price before you pay. See the Black Friday offer

Living with a chronic condition? Welltory Care is a paid support program built around the data you already collect, a community of people who share your symptoms, and medical experts who explain what the body is doing. Energy Lab is for women aged 18 to 65 with energy-limiting conditions such as ME/CFS, Long COVID, POTS and fibromyalgia; Migraine Lab is for women aged 18 to 65 living with migraine and opens on December 1, 2026. Read what Energy Lab is and what it costs and what Migraine Lab is and what it costs, or see how Care works in Welltory's Help Center. Care is peer support and education, not medical care.

How we made it

We based this guide on methods literature about individual responses (Atkinson and Batterham, 2015), personal variation in physiological responses (Zeevi and colleagues, 2015), the Agency for Healthcare Research and Quality user's guide to n-of-1 trials (2014), regression to the mean (Barnett and colleagues, 2005), placebo effects (the Cochrane review by Hróbjartsson and Gøtzsche and the open-label placebo trial by Kaptchuk and colleagues, 2010) and HRV monitoring in athletes (Plews and colleagues, 2013 and 2014). The meditation example draws on meta-analyses of meditation for stress and well-being (Goyal and colleagues, 2014), sleep (Rusch and colleagues, 2019) and HRV (Brown and colleagues, 2021), and on NCCIH safety guidance. The meditation example itself is a hypothetical illustration, not data from a real person. For the product section, we checked every function against the Welltory Help Center on October 22, 2026, and the annual price against welltory.com/plans on the same day.

Made with AI tools, then edited and fact-checked by the Welltory team. See our Editorial & AI policy.

Jane Smorodnikova, co-founder of Welltory, built the methodology behind how we read physiological data.

Written by Tatsiana Yashyna.