What is CBT-I, and why has your doctor not mentioned it?

Cognitive behavioral therapy for insomnia is what the clinical guidelines say to try before sleeping pills. The American Academy of Sleep Medicine gives it their only strong recommendation for chronic insomnia, and the European guideline reached the same conclusion again in 2023. It is a structured program, six weeks in most versions, and it is behavioral rather than pharmacological. Most people have never heard of it because writing a prescription takes two minutes and trained CBT-I clinicians are scarce.

The two parts that do the work

Dismantling studies, the trials that test the components separately, keep finding that two behavioral pieces carry the outcome. The 2023 European guideline names them outright: sleep restriction and stimulus control are the most active ingredients.

Stimulus control

Years of lying awake teach your brain that bed is where worrying happens. Stimulus control retrains the association: bed is for sleep and sex only. Awake for a while? Get up, do something boring, come back sleepy. It feels backwards, and it is load-bearing.

The sleep window

The counterintuitive core, and the one clinicians call time in bed restriction. If you sleep five and a half hours but spend nine in bed, your sleep goes shallow and broken. Shrinking time in bed to roughly what you already sleep concentrates it and rebuilds the pressure to fall asleep fast. When your sleep efficiency, time asleep divided by time in bed, holds above 85% for a week, the window widens by fifteen minutes.

Everything else in a CBT-I program, the sleep hygiene checklists, the relaxation audio, the mindfulness, is garnish. Pleasant, and not the reason it works. A program that leads with meditation content has the sequence backwards.

What six weeks feels like

• Week 0. A morning sleep diary establishes your baseline. Diary, not wearable: CBT-I runs on your own report.
• Week 1. One fixed wake time, seven days a week, and naps end. The wake time is the lever. Nobody can make you fall asleep on schedule, but you can always get up on one.
• Weeks 2 and 3. The sleep window opens. You will be more tired before you are better. This is where most people quit, and where support earns its keep.
• Weeks 4 and 5. The cognitive layer: scheduled worry time, and catching the catastrophic middle-of-the-night math, the "if I do not sleep now I will fail tomorrow" loop. This comes once your nights already show progress.
• After. A relapse plan. The habits keep working once the program ends, which is why CBT-I holds up better than pills at long-term follow-up.

Honest numbers

The efficacy figures you will read come from trials where a trained clinician delivered the program. In those trials most people improve meaningfully and some reach remission. That is the best result insomnia care has to offer, and it still means some people do everything right and stay stuck. Those cases belong with a sleep specialist, not another app. The largest pooled analysis found face-to-face courses of at least four sessions outperforming self-help, and a 2025 review of fully automated programs found the same gap. Software is the accessible option, not the strongest one.

What the trials move most is how fast you fall asleep and how much the night distresses you, more than total hours slept. Across 87 trials, total sleep time was the smallest effect measured. Anyone promising you extra hours is bluffing.

One thing you can skip: melatonin. The American Academy of Sleep Medicine recommends against it for chronic insomnia, whatever the gummy bottle says. Jet lag and circadian problems are a different question.

When CBT-I is the wrong first move

• Possible sleep apnea. Loud snoring, gasping, or a partner noticing pauses in your breathing. Get that assessed first. Cutting time in bed on top of untreated apnea helps nobody.
• Shift work and circadian disorders. A different problem, needing different tools.
• Under 18, pregnant, or managing bipolar disorder. The sleep window needs clinician supervision. CBT-I itself is recommended in pregnancy; it is the time-in-bed part that wants someone watching.
• Sleep medication. CBT-I works alongside it, but tapering is a decision for you and your prescriber, never an app.

How to get it

A trained clinician is the gold standard. The catch is supply: sleep-clinic waitlists run months. The self-help paths are the book Say Good Night to Insomnia, which runs about fifteen dollars, the VA's free Insomnia Coach app, which is a standalone five-week plan you can work through alone, and the paid app programs, which we compared honestly in Sleep Reset vs Stellar Sleep. If you are already working with a clinician, their companion app is CBT-i Coach, which is built to sit alongside sessions rather than replace them.

Where Still fits

Still is the CBT-I protocol as a six-week program on your phone, with a guide that answers at 3am for as long as your program runs and reasons from your own diary. $79, once, in the App Store. Nothing renews, nothing to cancel. A program built to end.

This page is education, not medical advice. Efficacy figures throughout refer to clinician-delivered CBT-I. Still is a self-guided program based on CBT-I techniques. It is not a medical device and does not diagnose or treat any condition.