Last reviewed: September 2026. General information, not medical advice — see our Disclaimer.
“Mental health app” covers products that have almost nothing in common. A mood tracker, a guided meditation library, a CBT course and a platform connecting you to a licensed therapist are four different things with four different evidence bases and four different risks.
Lumping them together is why people download the wrong one, conclude apps do not work, and stop looking.
The five categories
| Category | What it does | Reasonable for |
|---|---|---|
| Meditation and mindfulness | Guided practice, breathing, sleep audio | Everyday stress, wind-down routines |
| Mood and symptom tracking | Logs how you feel over time | Spotting patterns, preparing for appointments |
| Structured CBT programmes | Courses based on therapeutic techniques | Mild to moderate anxiety or low mood |
| Therapy platforms | Connects you to a licensed professional | When you actually need a clinician |
| Peer support | Community, shared experience | Isolation, feeling alone with something |
Only the fourth involves a licensed professional. The others are self-help tools, and describing them otherwise is the main way this category oversells itself.
What the evidence supports, by category
Meditation and mindfulness has the largest research base, with modest benefits for stress and wellbeing. The strongest studies cover structured multi-week programmes rather than apps, a distinction covered in our guide to whether meditation apps actually work.
Structured CBT programmes are the most promising category. Digital CBT has reasonable evidence for mild to moderate anxiety and depression, and adherence is the recurring limitation — completion rates in real-world use are considerably lower than in trials.
Mood tracking has little evidence of benefit on its own. Its genuine value is different: arriving at an appointment with eight weeks of data is more useful to a clinician than trying to remember how you have been.
Peer support helps with isolation and carries a specific risk: unmoderated communities can normalise harmful behaviour. Moderation quality is the thing to check.
The National Institute of Mental Health publishes a measured overview of what these tools can and cannot do in its guidance on technology and mental health treatment, including the point that most apps have not been independently evaluated.
The claim to be suspicious of
Any app describing itself as clinically proven without pointing to a specific published trial of that product. Evidence for meditation generally is not evidence for a particular app, and most apps in this category have never been independently tested.
The privacy problem is not theoretical
What you enter into a mental health app is among the most sensitive data you generate: mood, symptoms, substance use, relationships, sometimes thoughts of self-harm.
Many people assume this is protected by medical privacy law. Frequently it is not. Many consumer wellness apps fall outside those rules entirely. The U.S. Department of Health and Human Services sets out when an app is and is not covered in its guidance on HIPAA and health apps, and the distinction is narrower than most people assume.
Before entering anything personal:
- Find the “how we share your information” section. If it mentions advertising or marketing partners, assume the worst reading is the true one.
- Check whether a clinician is involved. Care delivered by a licensed provider is usually covered by stricter rules than a self-help app.
- Look for account deletion, and whether it deletes the data or only the login.
- Use a unique password and turn on two-factor authentication.
- Consider what you type. A mood score is less exposing than a free-text journal, and both are useful.
What none of them can do
This is the part that matters most, and it is usually in the small print.
Self-help apps are not appropriate as the primary response to a diagnosed condition requiring treatment, and they are not built for crisis. Most say so somewhere, and the disclosure is rarely prominent.
An app is not a crisis service
If you are having thoughts of harming yourself, contact a crisis line or emergency services, not an app. In the United States, call or text 988 for the Suicide & Crisis Lifeline. The U.S. Substance Abuse and Mental Health Services Administration also runs a free, confidential national helpline at 1-800-662-HELP, available 24 hours a day. Elsewhere, findahelpline.com lists services by country.
They also cannot diagnose. Some produce a score from a questionnaire and present it as a result; a screening questionnaire is a prompt to talk to someone, not a diagnosis. And they cannot change a situation that is causing the distress — an unmanageable workload needs a workload intervention, as covered in the warning signs of burnout.
How to choose
- Name the problem first. Stress, sleep, low mood, isolation, or needing a professional. Each points to a different category.
- If the answer is “I need a professional”, skip the self-help categories. Our guide on how to find a therapist covers that route, including employer programmes that cost nothing.
- Check the privacy policy before creating an account, not after.
- Prefer structure over libraries. A defined course beats a catalogue, because choosing is itself an obstacle.
- Give it 4 to 8 weeks, then judge against a goal you set in advance.
- Stop if it makes things worse. That happens, it is not a failure of effort, and it is worth mentioning to a clinician.
Why adherence is the whole problem
Across every category here, the limiting factor is not whether the technique works. It is whether anyone keeps using it.
Real-world engagement with mental health apps drops sharply in the first two weeks, and that gap between trial results and everyday use is the main reason the category underdelivers. Trials measure people who were enrolled, supported and followed up. You are none of those things.
Two implications worth acting on. A simpler tool you open daily beats a sophisticated one you abandon, so choose for the version of you who is tired rather than the one downloading it enthusiastically. And an app used alongside some form of human contact — a clinician, a group, a friend who knows you are doing it — tends to be sustained considerably longer than one used entirely alone.
Set the reminder before you need it
The point at which you most need the tool is the point at which you have least energy to remember it. A fixed time attached to something you already do daily is more reliable than intention.
Questions to ask before the free trial converts
- Have I opened it more than three times this week? If not, the subscription is the wrong problem to be solving.
- What specifically has changed? Measured against the goal you set at the start, not a general impression.
- Am I using it instead of something? An app that is quietly postponing a conversation with a clinician is costing more than the subscription.
- What is the annual price? These are usually quoted monthly and billed yearly.
- How do I cancel, and how do I delete my data? Two separate things, and the second is frequently harder.
The free options
Several public health bodies publish free, unbranded mental health resources — self-help materials, guided exercises and symptom information — with no account, no subscription and no data sharing. Crisis lines are free everywhere. Many employers include confidential counselling sessions at no cost, which is the most underused option in this entire article.
A paid app is buying structure and convenience. That can be worth it. It is worth knowing it is what you are buying.
Related guides
- Do meditation apps actually work? — the evidence for the largest category, in detail.
- How to find a therapist — when an app is not the right tool.
- Anxiety vs. stress — naming the problem before choosing a tool for it.
Frequently asked questions
Do mental health apps work?
It depends entirely on the category and the problem. Structured digital CBT has reasonable evidence for mild to moderate symptoms; mood tracking alone has little on its own. Most individual apps have never been independently evaluated.
Are mental health apps private?
Not always in the way people assume. Consumer wellness apps often fall outside medical privacy rules, and regulators have acted against companies sharing health data with advertisers. Read the sharing section before entering anything.
Can an app replace therapy?
No. Self-help tools can support someone with everyday stress or mild symptoms. They are not a substitute for a clinician where one is needed, and they are not built for crisis.
Which app should I start with?
Start from the problem rather than the product. Stress points to mindfulness, persistent low mood points to structured CBT or a professional, and isolation points to peer support or people rather than apps.
Are free mental health apps safe to use?
Free products are funded some other way, which makes the data question more pressing. Publicly funded resources from health bodies are generally the safest free option.