Why I put a client to bed later, not earlier

By Laura Kanadel, cert. Sleep Counsellour. 

A client came to me a while ago with a sleep diary that told a very ordinary story. Nine hours in bed each night. About five and a half of them actually asleep. Two to three hours spent lying there, awake, waiting.

Her instinct, like most people's, was to go to bed earlier. Give the problem more room to solve itself. It's the opposite of what she needed, and it's the opposite of what most people with this exact pattern need.

The number that mattered

Sleep researchers call it sleep efficiency: the percentage of time in bed that you're actually asleep.

Sleep efficiency = (total sleep ÷ time in bed) × 100

Healthy sleep sits above roughly 85 percent. My client was at 61. That gap, those two to three hours of nightly wakefulness, isn't a sign that she needed more opportunity to sleep. It's a sign that her brain had stopped treating the bed as a place for sleep at all, it had become a place of worry. 

The brain is extremely good at linking places to states of mind and her brina had begun lying awake, thinking, checking the time, getting frustrated, and the bed stopped signaling sleep. 

The method: CBT-I (cognitive behavioral therapy for insomnia)

The approach I used is cognitive behavioral therapy for insomnia, CBT-I. It's the treatment with the strongest evidence base for ongoing difficulty falling or staying asleep, and it's recommended internationally as the first line of treatment, ahead of sleep medication.

CBT-I rests on four pillars:

Sleep restriction. Time in bed is cut down to match the sleep a person is actually getting. This builds sleep pressure and shortens the time it takes to fall asleep.

Stimulus control. The brain relearns, deliberately, that the bed means sleep.

A fixed circadian rhythm, anchored by light. The body gets an unambiguous signal for when the day begins.

Cognitive work. Worry gets a scheduled place to happen that isn't the bed.

The name sounds harsher than the method feels. Nobody's sleep gets taken away. What gets removed is the time already being spent awake.

The sleep window

This is the core of the program. 

We set a fixed window, in her case six hours, based on what she was actually sleeping, not what she hoped to sleep. The rules inside that window don't bend:

• No time in bed before the window opens. Not to read, not to rest, not to watch something.
• A fixed wake time, seven days a week, including weekends and holidays.
• Out of bed within five minutes of waking, regardless of how the night went.
• No napping. Not even a short one in the afternoon. On its own, this single change often matters as much as the window itself.
• The bed is for sleep and nothing else.
• The phone sleeps outside the bedroom. An ordinary alarm clock does the job instead.

And one more, the rule that does the most work over time: if you're lying awake for roughly a quarter of an hour without falling asleep, you get up. Not by checking the clock, by feeling it. Go to another room, keep the light low and warm, do something quiet and slightly boring, sitting rather than lying down. Go back only when you feel real sleepiness: heavy eyelids, yawning, not just tiredness. Repeat as many times as needed. It isn't a setback. It's the exercise.

The window widens gradually, and only once sleep is stable, never before.

What to do before bed 

A sleep window that opens at midnight makes for a long evening, so the evening needs a shape, or it turns into nothing but waiting.

The architecture is simple: dinner at a normal hour, nothing to eat after except water. A fixed block of worry time (more on that below). Every screen off at a set hour, phone, television, laptop, tablet, no exceptions. A stretch of quiet, screen-free time: a podcast, a paper book, tidying up, a short walk. A warm bath or shower ninety minutes to two hours before the window opens. Lights dimmed through the house well before bedtime, lamps only, nothing overhead. 

Two pieces of that sequence are doing more than they look like they're doing. The screen cutoff matters because screens work against sleep twice over: the light suppresses melatonin, and the content keeps the brain engaged. The bath matters because of what happens after you get out. Warm water widens the blood vessels near the skin, and when you step out, core temperature drops faster than it otherwise would. 

Worry time

Most people I work with describe some version of the same thing: the mind gets busier the moment the lights go out. That's rarely a coincidence, and it's rarely really about the bed.

The brain doesn't let go of anything it considers unresolved. If it isn't given a chance to process something during the day, it will do it the moment you lie down. So we give it an appointment instead.

Fifteen to thirty minutes, at a table writing, never the sofa and never the bed. One side of the page for whatever's on your mind, the other side for the next concrete step you could take, not the full solution, just the next step. Close the journal and leave it on the table. If a thought surfaces later, in bed, the instruction is to tell it, gently, that it has an appointment tomorrow. If it won't wait, get up and write it down, then the quarter-hour rule applies.

What the day is doing to the night

Light. Twenty to thirty minutes outdoors, ideally between roughly 6:15 and 8:00am, no sunglasses, no window in between, every single day, including grey winter mornings. On days that's genuinely not possible, a light therapy lamp for the same duration. Morning light is the strongest signal the body has for setting its clock, and it's also what determines when melatonin starts rising later that evening. 

Movement. Around ten thousand steps a day, ideally with half of them before noon. Hard exercise finishes by late afternoon. After early evening, movement stays gentle, a short walk at most.

Stimulants. Last coffee by noon, and that includes tea, cola, energy drinks, and dark chocolate later in the day. No alcohol for the first month of the program: it makes falling asleep easier but suppresses REM sleep and drives more waking in the second half of the night. No nicotine.

Food. Breakfast within sixty to ninety minutes of waking, at a consistent time. Lunch by early afternoon. Dinner at a normal hour, enough to actually feel full.

The bedroom itself

Temperature matters more than most people expect. The body needs to lower its core temperature to fall asleep, and a warm room works directly against that. Aim for roughly 17 to 18°C: air the room out for ten minutes before bed even in summer, keep windows and curtains closed during the day when the sun is out. 

And the bed itself needs to go back to meaning one thing. No phone, no screen, no work, nothing in it before the window opens or after it closes. It's also worth checking something simple and easy to overlook: whether the pillow actually suits how someone sleeps. 

When it isn't only about sleep

My client also flagged restless legs, a sensation she described in her own words, appearing specifically in the evening and at rest. That combination, restless legs alongside a long time to fall asleep, isn't a coincidence.

Restless legs syndrome can be a real, separate driver of long time-to-sleep, one that sleep restriction alone won't resolve. It's also worth knowing that short-term sleep loss, which the early weeks of this program involve by design, can temporarily worsen the sensation. So this gets checked properly, in parallel, not instead of the program: bloodwork for iron status, specifically ferritin and transferrin saturation, since iron deficiency is the most commonly missed cause of restless legs.

This is the part I'd underline for anyone reading this and recognizing themselves: sleep coaching and medical evaluation aren't competing options. They run side by side.

What to actually expect

Nights one to three: staying awake until the window opens is often easier than people expect. Falling asleep tends to happen faster than usual.

Days three to ten: the hardest stretch. Noticeable tiredness through the middle of the day, a shorter fuse. This isn't a sign the program isn't working. It's sleep pressure building, which is exactly the point.

Weeks two to three: time to fall asleep starts dropping in a way that's genuinely noticeable. Many people are under thirty minutes by here.

Weeks four to six: sleep becomes more continuous, and this is where the window starts widening in earnest.

The three things that decide whether it works

Getting up at the fixed wake time, every day, without exception, especially after a bad night. This single habit is where the whole method stands or falls.

Not sleeping outside the window. No naps, no dozing on the sofa in the early evening, no extra hour at the weekend. Falling asleep in front of the television counts as sleep, and it quietly drains the sleep pressure the night depends on.

Not widening the window ahead of schedule, even when things are going well. Adjustments happen weekly, not daily.

Where this needs care

Sleep restriction is well studied and it works, but it does cause real, temporary tiredness in the early weeks. A few things follow from that: avoid driving or operating machinery on any morning where tiredness is significant, particularly in week one. The approach needs medical involvement and extra caution for anyone with bipolar disorder, epilepsy, untreated sleep apnea, sleepwalking, or during pregnancy. Anyone on sleep medication or other regular medication needs that factored in before starting.

And this matters most of all: sleep coaching is guidance, not medical treatment, and it doesn't replace either. Restless legs, in particular, belongs with a doctor, not with a sleep program alone.

Frequently asked questions

What is sleep restriction therapy? It's the core technique inside CBT-I, cognitive behavioral therapy for insomnia. Time spent in bed is reduced to closely match the amount of sleep a person is actually getting, which rebuilds sleep pressure and shortens the time it takes to fall asleep. The window widens gradually as sleep becomes more consistent.

What is a good sleep efficiency percentage? Sleep efficiency is the share of time in bed that's spent actually asleep: (total sleep ÷ time in bed) × 100. Above roughly 85 percent is considered good, above 90 percent is very good, and below 85 percent is generally considered poor and worth addressing.

How long does CBT-I take to work? Most people notice time-to-fall-asleep dropping by weeks two to three. Sleep tends to become meaningfully more continuous by weeks four to six. A full program typically runs six to ten weeks, adjusted weekly based on results, not on a fixed calendar.

Is sleep restriction therapy safe? It's well studied and effective, but it does cause real, temporary tiredness in the early weeks, so caution around driving or operating machinery matters early on. It requires medical involvement for anyone with bipolar disorder, epilepsy, untreated sleep apnea, sleepwalking, or during pregnancy, and it's guidance, not a replacement for medical treatment.

Can restless legs syndrome cause insomnia? Yes. Restless legs syndrome commonly shows up as difficulty falling asleep and can exist alongside, or independently of, other sleep issues. It's frequently linked to low iron stores, so a ferritin and transferrin saturation blood test is a reasonable first step, even when a standard blood count looks normal.