Two versions of me show up most weeknights: one is hunched over a campaign brief at full throttle, convinced she has just cracked the client's entire problem, and the other reads that same paragraph back the next morning and cannot remember what felt so brilliant about it. Only one of those two should be trusted with actual decisions. Figuring out how to shut the first one down before it burns another hour of sleep turned into the real work behind fixing my marketing burnout, and it is the piece of sleep optimization nobody warns you about when the job never quite clocks out.
I used to treat that late burst of clarity as proof I was built different, a night owl who did her best thinking once the house went quiet. Nodding off mid-pitch for a tech client cured me of that theory fast. What I was actually experiencing, most nights, was a stress response wearing a creative streak as a disguise.
The Second Wind Is a Cortisol Problem, Not a Productivity Win
Most of a marketing workday is passive stress: back-to-back calls, a triage inbox, a deck that needs one more revision by end of day. None of it asks your brain to actually build anything until the house is quiet and the Slack notifications finally stop. The 10 p.m. second wind is cortisol doing one final loop, and the goal is not to power through it but to cut the loop short before it locks you in for another hour.
It has nothing to do with talent and everything to do with timing relative to your own melatonin release, which is exactly why the fix is procedural, not motivational. A year into paying real attention to this, the pattern turned out to be almost boring: same trigger, same window, same workaround, every single time. None of it replaces an actual sleep specialist if the problem runs deeper than a bad evening habit. I am not a doctor, just someone who got tired of being tired and started paying attention to her own data.
The Thirty-Minute Window Before the Spike Locks In
Track your own evenings for a few nights and a pattern shows up fast: there is a specific half hour, usually somewhere between 10:15 and 10:45, when a screen is still open. If you are looking at one during that window, the spike is already happening; the plane has left the ground, and grounding it now costs a fight instead of a quiet redirect. The move is to be off screens before that window opens, not during it.
There's no negotiating with it once it starts. No "one last check," no "just replying to this one thing." The rule that actually holds is a hard stop at a fixed time — mine is 10 p.m. — treated the same way a launch deadline gets treated: non-negotiable, not a suggestion you renegotiate depending on mood.
What Does Nervous System Regulation Actually Require at 10 PM?
Not every fix earns a permanent place here. I tried drinking chamomile tea every night for two weeks and my tracker showed zero measurable change: a flat line where I expected at least a small dip in resting heart rate. Herbal tea was not the lever. Timing was.
What works instead is lower-stimulation and physical rather than mental: folding laundry, a slow and slightly boring skincare routine, anything that shifts the focus away from a brain still running numbers for tomorrow's meeting. The point is not relaxation exactly. It is giving the nervous system something present-tense and low-stakes to do instead of solving problems that can wait until morning. Cutting the actual cord — laptop closed, not just minimized — is its own separate discipline I have broken down in more depth elsewhere; here the point is only that the discipline has to start before the second wind, not during it.
Building an Evening Routine That Holds Under Deadline Pressure
A shutdown routine only proves itself if it survives a bad week, not just a calm one. Mine held through most of last year except for the genuinely worst stretches, which is honestly the whole test of whether a routine is real or just a good idea on paper. My friend Damien Ulrich, self-deprecating as always about his own terrible hours, laughed when I told him I now treat 10 p.m. like a hard deadline. He is not wrong to laugh. Neither of us has this fully figured out.
None of this works if the last coffee of the day landed at 4 p.m. and a deadline pushed dinner to 8; that's a caffeine-timing problem stacked on top of a cortisol one, and it deserves its own conversation rather than getting buried in this one. The routine itself is not complicated, but it does require rebuilding the hour before it, which is why I wrote a full breakdown of how I had to overhaul my after-work routine to protect my sleep separately. The shutdown moment does not work in isolation. The other piece is the kitchen. A second wind wants sugar, fast, and feeding it a bowl of cereal at 11 p.m. is close to signing up for a 2 a.m. bedtime; I covered how I finally started managing late-night cravings as a busy director in more detail elsewhere.
The Next-Morning Test for Every Midnight Idea
Every midnight brief gets one test now: read it again after coffee, before touching it further. If it still holds up, it stays. If it reads like three already-rejected ideas wearing a caffeine jacket, it goes in the trash, and that happens more often than the "genius" version of me from the night before would like to admit.
A former colleague, Rowan Ng, found the blog a while back and sent a message along the lines of: I cannot believe you actually wrote this down. He is a committed night owl who resents being one, and the part that landed for him was not the cortisol explanation; it was the thirty-minute window. Naming the window gave him something to actually act on instead of just feeling guilty about being awake.
These days I wake up most mornings a minute or so before my alarm, and there is no knot of anxiety sitting there waiting for me before I even open my eyes. That's the actual measure of whether any of this worked: not how brilliant last night's ideas felt in the moment, but how the next morning feels before the inbox opens.