I Tracked Everything for 30 Days — Here's What Actually Moved the Needle

// the_setup

I'm a software engineer. I solve problems for a living, and my ability to think clearly is directly correlated with my output and my income. So when I kept having inconsistent days — some mornings I'd be locked in for 4 hours straight, other mornings I couldn't focus through a single code review — I decided to stop guessing and start measuring.

The plan was simple: track every input that could plausibly affect cognition, measure cognitive output three times daily, and let the data reveal what actually matters.

What I tracked (inputs)

• Sleep: Bedtime, wake time, total hours, subjective sleep quality (1-10)
• Food: Every meal — what I ate, rough timing, approximate macros
• Caffeine: Exact timing and amount of every coffee, tea, or energy drink
• Exercise: Type, duration, intensity, and timing
• Supplements: What I took and when (creatine, omega-3, magnesium, vitamin D)
• Alcohol: Any consumption, type, and amount
• Screen time before bed: Hours of screen exposure in the 2 hours before sleep
• Water intake: Rough daily total

What I tracked (outputs)

• Focus score: 1-10 rating at 10am, 2pm, and 7pm
• Energy level: 1-10 at the same three checkpoints
• Deep work hours: Total hours of uninterrupted focused work per day
• Reaction time: A quick online reaction time test each morning (5 trials, averaged)
• Mood: 1-10 at end of day

Let me walk you through the five biggest discoveries — ranked by how much they surprised me.

// discovery_1: sleep consistency beat sleep duration

I expected total sleep hours to be the #1 predictor of next-day focus. It wasn't even close.

The strongest correlate with my morning focus score was sleep timing consistency — how close my bedtime and wake time were to my average. On days where I went to bed within 30 minutes of my usual time, my morning focus averaged 7.8/10. On days where bedtime shifted by more than an hour (even if total sleep was the same), morning focus dropped to 5.9/10.

Put differently: sleeping 7 hours from 10:30pm-5:30am consistently outperformed sleeping 8 hours from midnight-8am irregularly. The consistency of the sleep window mattered more than the duration.

This aligns with research on circadian rhythm stability. Your body's internal clock optimizes hormone release, body temperature, and neurotransmitter cycling based on expected sleep/wake times. When those times shift, the entire system desyncs. It's social jet lag — the biological equivalent of flying across time zones every weekend.

// discovery_2: the 48-hour food delay was real

I'd read about delayed food effects but was skeptical. Then my data showed it clear as day.

There was almost zero correlation between what I ate on a given day and my cognitive scores that same day. But when I offset the analysis by 48 hours — comparing Tuesday's food to Thursday's focus — patterns appeared that were unmistakable.

High-sugar days (days where I consumed significant refined sugar — dessert, candy, sugary drinks) correlated with a 1.8-point drop in focus scores two days later. Not the next day — two days later. The same-day effect was negligible.

High-protein, moderate-fat meals showed the opposite pattern: focus scores were approximately 1.2 points higher 48 hours after high-protein days compared to my average.

The most dramatic finding: three days after drinking more than two alcoholic drinks, my average focus score was 2.4 points lower than baseline. Not the hangover day — that was only -0.8. The real hit came on day three, when I felt "fine" but my data said otherwise.

// discovery_3: morning exercise was a cheat code

I alternated between morning workouts (7am), evening workouts (6pm), and rest days throughout the experiment. I expected exercise timing to have a moderate effect on focus. The actual difference was massive.

Morning exercise days: Average 10am focus score of 8.2/10. Average deep work hours: 4.8 hours.

Evening exercise days: Average 10am focus score of 6.9/10 (next morning). Average deep work hours: 3.6 hours.

Rest days: Average 10am focus score of 6.5/10. Average deep work hours: 3.2 hours.

Morning exercise — specifically, 30-45 minutes of moderate-intensity activity (lifting or a run) — boosted my cognitive output by roughly 50% compared to rest days. Not a subtle effect. Not a marginal gain. A genuine, measurable 50% increase in focused output.

The mechanism is well-documented: exercise increases BDNF (brain-derived neurotrophic factor), which enhances neuroplasticity, learning, and focus. It also elevates norepinephrine and dopamine for several hours post-workout. Morning exercise gives you those elevated levels precisely when you need them — during peak work hours.

// discovery_4: caffeine timing was more impactful than caffeine amount

I experimented with caffeine timing throughout the month. Some days I had coffee at 6am (immediately on waking). Other days I delayed until 9:30-10am. The total amount stayed the same — roughly 200-300mg per day.

Days where I delayed my first coffee to 9:30am or later showed consistently higher afternoon focus scores — averaging 7.1/10 at the 2pm checkpoint versus 5.4/10 when I drank coffee immediately on waking. That's a 1.7-point difference in afternoon performance from the same amount of caffeine, just shifted by 3 hours.

Even more interesting: on delayed-caffeine days, my morning focus before the coffee was better than I expected — averaging 6.4/10. The mornings weren't the low-energy hell I'd feared. It turns out that the cortisol awakening response does provide genuine alertness if you stop masking it with caffeine.

The biggest revelation was my caffeine cutoff. Days where my last caffeine was before noon correlated with 0.6 points higher sleep quality than days where I had afternoon coffee. And that improved sleep quality cascaded into the next morning's performance. It's a compounding effect — better caffeine timing → better sleep → better next-day performance → less need for caffeine.

// discovery_5: the things that didn't matter

Just as revealing as what worked was what didn't. Several things I assumed were important showed no measurable correlation with my cognitive scores:

Water intake: I tracked daily water consumption ranging from 4 glasses to 12 glasses. Unless I was actively dehydrated (rare), there was no correlation between water intake and focus scores. The "drink 8 glasses" advice had zero signal in my data. If you're drinking when you're thirsty, you're probably fine.

Specific supplement timing: Taking my supplements (creatine, omega-3, vitamin D, magnesium) in the morning versus evening made no measurable difference. The supplements themselves showed modest effects over the full month, but timing was noise.

Meal frequency: I tested 2 meals/day, 3 meals/day, and occasional intermittent fasting (16:8). No consistent effect on cognitive scores. The content of meals mattered. The timing and frequency didn't — at least for me.

Work environment changes: Standing desk vs. sitting, background music vs. silence, different lighting — none of these showed a meaningful pattern against my focus scores. The internal inputs (sleep, food, exercise, caffeine) dominated the external ones by a wide margin.

// what_changed_after

Based on 30 days of data, I made four changes to my daily routine:

1. Non-negotiable sleep window. I set a hard bedtime of 10:15pm (±15 min) and wake time of 5:45am. Weekends too. Social jet lag is real and I was doing it to myself every Friday and Saturday. The consistency matters more than the occasional extra hour.

2. Morning exercise, every working day. 30 minutes minimum. Doesn't have to be intense — even a brisk walk shows most of the benefit. But it happens before I sit down at my desk, period. The data was too clear to ignore.

3. Delayed caffeine, early cutoff. First coffee at 9:30am. Last caffeine by noon. Same total intake, radically different performance curve throughout the day. Afternoons went from my weakest period to consistently productive.

4. Sunday meal prep focused on protein. Since the 48-hour food delay was so pronounced, I started thinking about Tuesday's focus on Sunday. High-protein, low-sugar meal prep for the week ahead. My weekday cognitive average went up by nearly a full point.

The things I didn't change: water tracking (stopped — it was noise), supplement timing (whenever, doesn't matter), meal frequency (stuck with what felt natural).

// what_i_learned_about_tracking

The meta-lesson from this experiment: intuition is a terrible predictor of what actually matters.

Before tracking, I would have bet money that sleep duration, hydration, and supplements were my top three levers. They weren't even close. The actual levers — sleep consistency, morning exercise, and delayed food effects — were things I either undervalued or didn't know about at all.

This is why tracking matters. Not because it's fun. Not because you enjoy logging meals. But because your brain is bad at identifying the variables that drive your own performance. You need data to override your assumptions.

The hardest part was the first two weeks. Logging everything felt tedious. But once patterns started emerging — once I could see that Monday's pizza reliably predicted Wednesday's brain fog — it became genuinely interesting. I wasn't just tracking; I was reverse-engineering my own biology.

// the_bottom_line

Thirty days of comprehensive tracking taught me more about my own cognitive performance than thirty years of guessing. The discoveries were counterintuitive, personal, and actionable.

I don't track at the same intensity anymore — I don't need to. The habits are built and the major patterns are known. But every few months I'll run a focused experiment when something feels off or I want to test a new variable.

If you're a knowledge worker and you've never done this, you're leaving performance on the table. You have no idea what's actually driving your good days and your bad days. You think you know. I thought I knew too.

I was wrong about almost everything. The data fixed that.