Andrew Huberman's caffeine delay protocol has probably reshaped more morning routines than any other single piece of health content in the past five years. The recommendation: wait 90 to 120 minutes after waking before your first coffee. The claimed benefit: avoid the mid-afternoon energy crash, feel more alert throughout the day, and sleep better at night.
It is one of those pieces of advice that sounds mechanistically plausible and has been shared so widely that it feels like settled science. The reality is more nuanced. The underlying chronobiology is real. The specific 90-minute prescription, applied universally, is not well-supported by direct evidence. And the distinction matters if you are trying to understand what you are actually doing when you follow this protocol.
The Theory: Adenosine and the Cortisol Awakening Response
To evaluate the caffeine delay protocol, you need to understand two separate mechanisms that Huberman combines into a single recommendation.
Adenosine: Your Sleep Pressure Molecule
Adenosine is a byproduct of cellular energy metabolism. It accumulates in your brain throughout the day and binds to adenosine receptors, progressively increasing your subjective sense of tiredness -- a process called sleep pressure. Caffeine works primarily by blocking adenosine receptors, not by increasing energy. You are not adding energy; you are suppressing the signal that tells you you are tired.
The relevant question is: what happens to adenosine while you sleep?
During sleep, your glymphatic system -- the brain's waste clearance network -- clears accumulated adenosine. By the time you wake, adenosine levels are at their lowest. The sleep pressure clock has been reset.
The Huberman argument is that drinking caffeine immediately after waking blocks adenosine receptors when adenosine levels are already low. When the caffeine clears (roughly 5-7 hours later), the adenosine that accumulated during the morning suddenly floods the now-unblocked receptors -- producing the characteristic afternoon energy crash.
If you wait until adenosine has begun to re-accumulate naturally, the argument goes, you are using caffeine at a moment when it is actually doing something -- blocking real adenosine buildup rather than theoretically suppressing an already-cleared signal.
Adenosine clearance during sleep is not complete for everyone. Sleep debt, poor sleep quality, and sleep disorders all affect how much adenosine is cleared overnight. People with poor sleep may wake with measurable adenosine levels that partly explain the immediate need for caffeine they feel.
The Cortisol Awakening Response
The second mechanism is the cortisol awakening response (CAR). Cortisol, often framed as a "stress hormone," also serves as your primary arousal and alertness signal. It follows a strong diurnal rhythm, peaking in the first 30-45 minutes after waking -- a phenomenon well-documented in chronobiology research.
This CAR peak is significant. Studies have found it to be approximately 50-100% above baseline cortisol levels, providing a natural, endogenous alertness boost in the early morning. It is, essentially, built-in wakefulness machinery.
The argument for delaying caffeine: your cortisol is already doing the job in the first 30-60 minutes post-waking. Adding caffeine on top of a cortisol peak does not meaningfully increase alertness -- it just potentially blunts your cortisol response through tolerance mechanisms and wastes the caffeine window.
Related: Want to put this into practice? Try our Sleep Score Calculator to get started, and check out Best Energy Supplements for Men 2026 for more context.
What the Evidence Actually Shows
Here is where intellectual honesty requires a step back.
The mechanistic framework described above is grounded in real science. Adenosine clearance is real. The cortisol awakening response is well-documented and reproducible. Caffeine's mechanism of action via adenosine receptor antagonism is not in dispute.
What does not exist is a well-designed randomized controlled trial directly testing whether waiting 90 minutes after waking to consume caffeine reduces afternoon energy crashes, improves performance, or affects sleep quality compared to immediate post-waking consumption.
The 90-minute figure specifically appears to originate from extrapolation: cortisol peaks at roughly 30-45 minutes post-waking, and waiting until it has declined (assumed to take 60-90 minutes) is positioned as the right window to introduce caffeine. This is logical inference from the biology -- not direct experimental evidence.
A 2024 review in Frontiers in Nutrition examined caffeine timing research and found that while overall daily caffeine timing affects sleep onset and quality, the specific "delay post-waking" hypothesis has not been tested in controlled trials. Most evidence on caffeine and sleep focuses on avoiding caffeine in the 6-8 hours before bed -- a much more robustly supported intervention.
Individual Variation Makes This Complicated
Even if the mechanism is sound on average, the optimal delay is likely highly individual. Several factors influence it:
Chronotype. Morning chronotypes (natural early risers) have an earlier and often stronger cortisol awakening response than evening chronotypes. Someone who naturally wakes at 5 AM may clear their CAR peak faster than someone who wakes at 9 AM.
Sleep quality. Poor sleep partially blunts the CAR. Someone coming off a bad night may have a weaker cortisol peak and lower adenosine clearance -- meaning the calculus changes entirely.
Caffeine metabolism. CYP1A2 enzyme variants significantly affect how quickly individuals metabolize caffeine. Fast metabolizers clear caffeine in 3-4 hours; slow metabolizers in 7-9 hours. The appropriate delay and dose are different for each.
Habitual intake. Chronic caffeine consumers develop meaningful receptor tolerance. The adenosine receptor dynamics in a daily two-cup drinker are different from a non-habitual user.
Be the first to try Prova
We're building an app to track whether caffeine timing actually works. Join the waitlist.
What Is Actually Well-Supported
Stepping back from the 90-minute debate, two caffeine-related interventions have substantially stronger evidence:
1. Avoid caffeine in the 8-10 hours before bed. This is the most evidence-backed caffeine timing intervention. A 2013 study by Drake et al. in the Journal of Clinical Sleep Medicine found that caffeine consumed even 6 hours before bedtime significantly reduced sleep duration and quality as measured by polysomnography. Most people underestimate the half-life of caffeine (roughly 5-6 hours in healthy adults, longer in slow metabolizers) and consume afternoon coffee that is still active at bedtime.
2. Hydrate before caffeine. After 7-8 hours of sleep, you are mildly dehydrated. Mild dehydration (as little as 1-2% body weight) impairs cognitive performance and mood. Drinking water before caffeine addresses a real and immediate physiological state.
The delay protocol may be beneficial for some individuals, but its effect size is likely smaller than simply not drinking coffee after 2 PM.
How to Run Your Own Experiment
This is a perfect candidate for personal experimentation because the outcome you care about -- afternoon energy, subjective alertness, sleep quality -- is measurable.
A simple N=1 protocol:
Week 1: Drink caffeine at your usual time (immediately or within 30 minutes of waking). Track afternoon energy (1-10 scale at 2 PM and 4 PM), sleep quality (subjective or via wearable), and time to fall asleep.
Week 2: Delay caffeine to 90 minutes post-waking. Same tracking.
Week 3: Try 60 minutes. Compare all three.
The variables to control: caffeine dose, total daily intake, last intake time, sleep duration, and alcohol consumption. Keep those constant and the signal from timing differences becomes detectable.
If you use a wearable that tracks HRV, resting heart rate, or sleep staging, you already have objective data to layer on top of subjective energy ratings. Even two weeks of data can reveal whether timing changes are producing measurable physiological effects for you.
Pros
- +Mechanistically plausible -- adenosine clearance and cortisol awakening response are both real phenomena
- +Low cost to try -- requires no supplements or equipment
- +May reduce afternoon energy crashes if the mechanism applies to you
- +Forces morning hydration and sunlight exposure before caffeine, both independently beneficial
- +Compatible with intermittent fasting protocols
Cons
- -No direct RCT evidence for the 90-minute delay specifically
- -Highly individual -- chronotype and sleep quality significantly affect optimal timing
- -Difficult to maintain in social or work contexts (morning meetings, early calls)
- -People with disrupted sleep may actually need early caffeine to function safely
- -May not be necessary if you are already avoiding late-afternoon caffeine