Which Coping Skills Actually Work for You (and How to Find Out)

You have six breathing apps on your phone. You have tried box breathing, cold water on your face, a gratitude list, a walk, and the one your therapist keeps mentioning. Some days you feel better afterward. You still could not say out loud which of them does anything for you. That question has an answer, and two weeks is usually enough to get it.

Why the lists do not help anymore

Search for coping skills and you get catalogs. Lists of 13. Lists of 25. Lists of 100. They are not wrong, and most of what sits on them has research behind it. What they measure is what happens on average, across hundreds of people, and you are not an average.

Medicine has a name for the gap. Kravitz, Duan and Braslow called it the trouble with averages in a 2004 paper in The Milbank Quarterly, where they argued that one average effect can hide substantial benefit for some people, very little for many others, and harm for a few. Group results tell you what is worth trying. They cannot tell you what worked in your kitchen on a Tuesday.

So the useful question is not which coping skills work, but which ones work for you. The only way to answer that is to check.

The one variable rule

The method, in five steps:

1. Pick one skill and change nothing else.
2. Pick one narrow measure, rated on the same scale at the same time every day.
3. Write down what you predict will happen, with a number, before you start.
4. Run it for two weeks without judging it in the middle.
5. Compare week one to week two on averages, not on your best day.

None of this is new. Cognitive behavioral therapy has done a version of it for decades under the name behavioral experiment: take a technique or a belief, set up a small test in real life, and see what comes back. The Oxford Guide to Behavioural Experiments in Cognitive Therapy, edited by Bennett-Levy and colleagues in 2004, collects more than 200 of them. Medicine has a stricter cousin, the N-of-1 trial, described by Guyatt and colleagues in a 1986 paper in the New England Journal of Medicine: one patient, one question, repeated measurement.

Your version at home will not be blinded and it is not a clinical trial. The logic still transfers.

Pick one thing, not four

The usual mistake is starting on Monday with breathing, journaling, an earlier bedtime and a walk at lunch. Two weeks later something has shifted and you have no idea which piece did it. Or nothing has shifted, and you write off all four when one was quietly working and the other three were eating its time.

One skill. The next one waits.

Two weeks is the minimum

A single week gives you five or six data points, and one bad Wednesday swallows the picture. Two weeks leaves room for a normal-sized disaster to average out. If the thing you are testing is slow by nature, like sleep or a regular movement practice, give it four.

Choose the measure before you start

Rate the thing you are trying to change, not your mood in general

"How was my mood today" is too broad to move. If you are testing a breathing exercise for anxiety before meetings, rate anxiety before meetings. If you are testing an evening walk for sleep, rate how long it took you to fall asleep. A narrow measure moves when the thing you are testing moves. A broad one gets drowned out by everything else that week.

Same scale, same time of day

0 to 10 is fine. Whatever you pick, keep it identical for the full two weeks and answer it at the same point in your day. Ratings made from memory drift.

Rating tonight what you felt on Thursday morning is a memory test, not a measurement.

Wellbeing check-ins
Rate the one thing you are testing in under a minute, at the same time every day. Try it free in MindSync →

Write the prediction down first

Before day one, finish this sentence: if this works, in two weeks my rating will be around this number instead of that one. Put an actual figure in it. Save it somewhere you cannot quietly edit later.

The step feels pointless and it is the one that saves the experiment. With no number written in advance, you will read whatever those two weeks produced as confirmation of what you already believed. Written first, a prediction gets met or missed.

Read the result honestly

What a real signal looks like

Compare week one with week two, average against average, not your best day against your worst. A real signal shows up as a shift in the typical day: your usual 6 becomes a usual 4, and it holds across most days rather than one. Small and consistent beats large and isolated.

What noise looks like

One good week settles nothing. Weeks vary on their own for reasons that have nothing to do with what you are testing. A deadline landed or it did not. Someone was kind to you. You slept. If your average barely moved and one spectacular Tuesday is doing all the work, what you found is a Tuesday.

One more complication. The act of tracking something can change it. Korotitsch and Nelson-Gray reviewed the evidence in a 1999 paper in Psychological Assessment: reactivity is a documented property of self-monitoring, so part of any improvement you record may belong to the recording rather than the skill. Measuring is still worth doing. Just stay modest about small effects, and keep tracking after the test ends to see whether the gain survives.

When to stop, when to extend

Clear improvement means you keep the skill, make it boring, and go test the next thing. Nothing at all means you drop it: two weeks spent, one fact bought. An ambiguous result usually means the measure was too broad rather than that the answer is out of reach, so sharpen it or run another two weeks.

Mood & focus-area trends
Two weeks of ratings next to each other, so a good day stops passing for progress. Try it free in MindSync →

Keeping two weeks of numbers side by side is easy to describe and tedious by hand. That is the part we keep working on: a way to see if the small things actually help you.

A worked example, start to finish

Here is the whole thing on one page. The numbers are invented to show the shape of the method, not results from a study.

Maya has been in therapy for eight months, mostly about work. Her therapist suggested five minutes of slow breathing before her Monday leadership meeting. She tried it twice, felt nothing much, and let it go.

• The one variable: five minutes of slow breathing before meetings, nothing else new for two weeks.
• The measure: anxiety in the 10 minutes before a meeting, 0 to 10, logged as the meeting starts rather than that evening.
• The prediction, written on day zero: my average pre-meeting rating drops from about 7 to about 5.
• Week one: 7, 8, 6, 7, 7. Average 7.
• Week two: 6, 5, 6, 4, 6. Average 5.4.

Prediction roughly met, and the improvement turns up on four days out of five instead of in one dramatic drop. Maya keeps the exercise, tells her therapist how the two weeks went, and picks the next thing to test.

The version where it fails is worth as much. Had week two come back 7, 6, 8, 7, 7, the honest read is that five minutes of breathing does little for her anxiety before meetings, which is good to know before a year of feeling guilty about skipping it.

One boundary matters here, and it is not negotiable. If a therapist or a doctor recommended something, two weeks of your own ratings is not a reason to stop it on your own. Take the numbers to the appointment and decide together. This method is for choosing between things worth trying, not for overruling the people treating you.

Not sure what to test first? Slow breathing and grounding have the most evidence behind them, and we went through the detail in the regulation exercises worth testing first.

The same logic scales up. Point it at your treatment as a whole instead of one technique and you get how to know if therapy is working, measured instead of guessed. The rest of what the app keeps track of is on the features page.

Running the experiment

• Wellbeing check-ins - The daily rating that takes under a minute.
• Homework & to-dos - What your therapist suggested, waiting on your home screen instead of in your memory.
• Weekly & monthly summaries - The week read back to you, so you compare notes instead of impressions.

Frequently asked questions

How long should I try a coping skill before giving up on it?

Two weeks of daily ratings is the practical minimum, because a single week can be ruined by one bad day. Slower changes like sleep or a movement routine need about four weeks. Set the end date before you start, so the decision does not get made on the day you happen to feel worst.

What if nothing seems to help?

Most of the time the measure was too broad, or too many things changed at once. Try again with one technique and one narrow rating. If several honest attempts come back flat, that is information for your therapist rather than a verdict on you, and it is worth bringing to the next session.

Can I test two things at once?

You can, but then you will not know which one did the work, and that was the point of testing. If you are impatient, run them back to back: two weeks on the first, two weeks on the second, same rating and same scale both times so the results are comparable.

How do I know if it is the skill or just a good week?

Look at the average across all 14 days rather than your best day. A real effect shows up on most days, not one. If a single dramatic day is carrying the whole difference, run another two weeks and see whether the pattern repeats or that Tuesday was just a good Tuesday.

Should I tell my therapist what I am testing?

Yes, and preferably before you start. Your therapist can tell you whether the rating makes sense, what a realistic timeframe looks like, and whether the technique fits what you are working on. It also hands you something concrete to open the next session with, instead of reconstructing two weeks out loud.