Real studies. Not vibes.

A 2020 meta-analysis by Grgic et al. pooling 21 studies found caffeine reliably increases muscular strength and power output. For an 80kg lifter that's roughly 240–480mg (2–4 cups of coffee) taken about 45–60 minutes before training, when blood levels peak. Skip it if you train in the evening, since caffeine's ~5–6h half-life will wreck your sleep, which is where the actual muscle is built.

Grgic, J. et al. (2020). Br J Sports Med: caffeine & strength meta-analysis.

Muscle adapts to demands placed on it (Schoenfeld, 2017). If the demand never rises, neither does the muscle. Aim to beat your last session by a small margin: one more rep, 2.5kg more, or one more set. Logging every set is what makes this possible; memory is unreliable. Even 1% per week compounds to ~68% stronger in a year.

Schoenfeld, B.J. (2017). J Strength Cond Res.

A meta-analysis (Schoenfeld et al., 2016) found training a muscle 2×/week produced significantly more hypertrophy than 1×/week even when total volume was equal, thanks to more frequent muscle-protein-synthesis spikes. Upper/lower or push/pull/legs splits naturally hit everything at least twice.

Schoenfeld, B.J. et al. (2016). Sports Med: frequency meta-analysis.

Schoenfeld et al. (2016) found 3-minute rest intervals produced greater strength and hypertrophy than 1-minute rests, because longer recovery lets you keep your reps and load high across sets. Rushing big lifts like squats and rows just bleeds performance. Save the short rests for isolation/pump work.

Schoenfeld, B.J. et al. (2016). J Strength Cond Res: rest intervals.

The reps that drive hypertrophy are the hard ones near the end of a set. Stopping with 0–3 reps in reserve (RIR) captures almost all the stimulus while managing fatigue. If you could have done 6 more reps, the set did little. Push working sets genuinely hard: you should be uncertain you'd get the next rep.

Effort/proximity-to-failure literature (e.g. Refalo et al., 2022).

Leproult & Van Cauter (2011) showed one week of 5-hour nights reduced testosterone 10–15%. Growth hormone is mostly released in deep sleep, and recovery (not the workout itself) is when you adapt. Treat 7–9 hours as a non-negotiable training variable, especially in a hard block or a cut.

Leproult, R. & Van Cauter, E. (2011). JAMA.

Morton et al. (2018) found total daily protein is the strongest dietary predictor of muscle gain, roughly 1.6–2.2g/kg for someone training. The post-workout "anabolic window" is real but wide (several hours), so don't stress timing. Spread it across 3–5 meals of ~0.4g/kg each to keep muscle protein synthesis elevated through the day.

Morton, R.W. et al. (2018). Br J Sports Med.

An American Heart Association study found that just increasing fibre to ~30g/day produced weight loss comparable to a more complicated prescribed diet. Fibre slows digestion, boosts satiety, feeds gut bacteria, and improves insulin sensitivity. Easiest wins: beans, oats, berries, and leaving the skin on fruit and potatoes.

Ma, Y. et al. / AHA: simple fibre-only intervention.

In Hall et al. (2019), a tightly controlled NIH trial, people ate ~500 kcal/day more on an ultra-processed diet than a whole-food diet matched for calories, sugar, fat, and fibre (and gained weight) despite the meals being equally palatable. Processed food is engineered to be eaten fast and over-consumed. Build meals around single-ingredient foods and the calories largely take care of themselves.

Hall, K.D. et al. (2019). Cell Metabolism.

Dennis et al. (2010) found adults who drank 500ml of water before each meal lost significantly more weight than those who didn't: the stomach distension blunts appetite so you eat less. Free, zero-effort, and it doubles as a hydration cue. Make it a habit before lunch and dinner.

Dennis, E.A. et al. (2010). Obesity.

Your body compensates poorly for calories from drinks: a 250 kcal soda or juice doesn't make you eat 250 kcal less later, unlike solid food. Sugary drinks are one of the easiest places to silently consume hundreds of calories. Default to water, black coffee, or tea, and eat your fruit rather than juicing it.

DiMeglio & Mattes (2000); Pan & Hu (2011).

High-protein breakfasts increase satiety hormones and reduce evening snacking and cravings compared to high-carb or skipped breakfasts (Leidy et al., 2013). Aim for 30g+ at your first meal (eggs, Greek yoghurt, or a shake) to steady appetite and make hitting your daily protein target far easier.

Leidy, H.J. et al. (2013). Am J Clin Nutr.

Sleep regularity, going to bed and waking at the same clock time every day, is one of the strongest levers on sleep quality, and the day-to-day consistency of your sleep-wake timing predicts health outcomes even better than total hours (Windred et al., 2024). When you start a new schedule and hold it strictly, your circadian system re-aligns in stages. Around Day 3 (the "survival shift"), homeostatic sleep pressure starts pulling you toward the new times, though mornings still feel groggy. By about Day 14 (the "physiological lock"), your melatonin, cortisol and core-temperature rhythms have re-synced, so you begin waking just before the alarm and feel alert in the day. Roughly Day 90 (the "neurological habit") is when the schedule runs on autopilot rather than willpower. The catch is the weekend exception: in the early weeks the rhythm is fragile, and sleeping in or staying up late for even a couple of days causes "social jet lag" that can reset your progress toward Day 1. Your circadian system shifts only about an hour or two per day, so adjust in 15-30 minute steps and anchor on a fixed wake time. Matteca tracks this for you as a consistency streak and awards the three milestone badges, removing them if you break the streak.

Sleep Foundation; Harvard Health; Johns Hopkins Medicine; Windred et al. (2024), SLEEP.

Gooley et al. (2011) showed exposure to ordinary room light before bed suppressed melatonin and shortened its duration. In the last 1–2 hours before sleep, dim the lights, switch to warm/low lamps, and cut overhead lighting. Get phone and laptop screens off your face or on night mode. Darkness is the single strongest signal that tells your brain it's time to sleep.

Gooley, J.J. et al. (2011). J Clin Endocrinol Metab.

Digesting a meal raises metabolism and core body temperature, but falling asleep requires your core temp to drop. Eating close to bedtime also worsens overnight blood-sugar control and can trigger reflux. Aim to finish your last real meal ~3 hours before bed; if hungry, keep it small and light.

Crispim, C.A. et al. (2011); circadian metabolism research.

Harvard sleep research shows a steady wake time (±30 min) is the most effective lever for sleep quality. Your circadian clock uses wake time, not bedtime, as its main anchor; "social jetlag" from sleeping in on weekends scrambles melatonin and cortisol rhythms. Pick one wake time and hold it, and bedtime will start to fall into place on its own.

Czeisler, C.A. et al. Harvard Sleep Medicine.

Morning light exposure advances your circadian rhythm, boosts daytime alertness, and helps you feel sleepy at the right time that night. Aim for 10 minutes of outdoor light within an hour of waking (more if it's overcast, since indoor light is far dimmer than it looks). It's the natural counterpart to dimming lights at night.

Circadian phototherapy literature (e.g. Blume et al., 2019).

Drake et al. (2013) found caffeine taken even 6 hours before bedtime reduced total sleep by over an hour, and people often didn't notice the impairment. With a ~5–6 hour half-life, an afternoon coffee is still half in your system at midnight. Set a cutoff around early afternoon (e.g. 2pm) and switch to decaf or tea after.

Drake, C. et al. (2013). J Clin Sleep Med.

Sleep onset depends on your core body temperature falling. A cooler room (~18°C / 65°F) supports that drop, which is why a hot bedroom makes you toss and turn. A warm shower 1–2 hours before bed helps too: it pulls blood to the skin so your core cools faster afterward.

Harding, E.C. et al. (2019). Front Neurosci: thermoregulation & sleep.

Hammering your brain with high-stimulation dopamine hits (social media, news, notifications) first thing raises your baseline so that ordinary focused work feels boring by comparison. Protect the first hour: no phone, no feeds. Start with something low-stimulation (water, light, a quick plan) and your hardest task will feel far easier to start because nothing has out-competed it yet.

Dopamine baseline / reward-prediction research (Lembke, 2021).

Research on deliberate practice (Ericsson) and Newport's Deep Work show top performers concentrate intense focus into a few protected blocks rather than grinding all day. Schedule 1–2 blocks of ~90 minutes for your most cognitively demanding work, single-tasked, and treat the rest of the day for admin and recovery. Quality of attention beats hours logged.

Newport, C. (2016). Deep Work; Ericsson, K.A.

APA-reviewed research found mental task-switching can cost up to 40% of productive time, and each switch leaves an "attention residue" that lingers on the previous task. Batch similar work, close irrelevant tabs, and finish one thing before opening the next. What feels like multitasking is mostly the tax of constant context-switching.

American Psychological Association: multitasking & switch costs.

Ward et al. (2017), the "Brain Drain" study, found that simply having your smartphone visible reduced available cognitive capacity, even when it was off. The fix isn't willpower; it's distance. During deep work, put the phone in another room or a drawer. Out of sight genuinely frees up mental bandwidth for the task.

Ward, A.F. et al. (2017). J Assoc Consumer Research.

Doing your most important or dreaded task first ("eating the frog") means it gets your sharpest attention and removes the background dread that drags on the rest of your day. Decide the single most important task the night before, and start the day on it before email or chat can hijack your agenda.

Productivity & self-regulation research (Baumeister, ego-depletion).

Ariga & Lleras (2011) found that brief breaks during a long task prevent the steady decline in focus that otherwise sets in. Structures like Pomodoro (25/5) or 52/17 work because they force recovery before you burn out. Step away from the screen, move, look at something far away, then come back fresh rather than grinding to diminishing returns.

Ariga, A. & Lleras, A. (2011). Cognition.

The most reliable way to save is to remove the decision. Set up an automatic transfer to savings or investments that fires the day your income lands, so you budget on what's left rather than hoping there's something left to save. This "pay yourself first" rule (Bach's Automatic Millionaire) beats willpower because the saving happens before lifestyle spending has a chance to absorb it.

Bach, D. The Automatic Millionaire; behavioural finance.

$5/day × 365 × 20 years at an 8% average return compounds to roughly $79,000, and the last 5 years add more than the first 15 combined. The point isn't to never buy coffee; it's to notice which small recurring expenses you don't actually value, and redirect those to investing. Audit your subscriptions and daily habits, not your occasional treats.

Compound interest math; historical S&P 500 ~8%/yr.

A clean default: 50% of after-tax income to needs (rent, food, bills), 30% to wants, 20% to saving and paying down debt. It's not sacred (adjust the ratios to your goals) but it gives you a target to compare reality against. If savings is well under 20%, the wants bucket is usually where it leaked.

Warren, E. & Tyagi, A. All Your Worth.

Impulse spending rides a spike of excitement that fades fast. Imposing a 24-hour (or for big items, 30-day) wait before non-essential purchases lets that spike pass, and most "must-haves" quietly lose their pull. Keep a wishlist; if you still want it after the wait and it fits the budget, buy it guilt-free.

Behavioural economics: hyperbolic discounting / cooling-off.

SPIVA data consistently shows the large majority of actively managed funds underperform their benchmark index over 10–15 years, largely because of fees. A broad, low-cost index fund captures the market's return cheaply and removes the need to pick winners. Boring and automatic beats clever and expensive for almost everyone.

S&P SPIVA Scorecards; Bogle, J.: index investing.

Before aggressive investing, hold 3–6 months of essential expenses in accessible cash. Without it, a job loss or surprise bill forces you onto credit cards at 20%+ interest, wiping out years of investment gains. The emergency fund isn't about returns: it's insurance that keeps the rest of your plan intact when life happens.

Standard personal-finance consensus (Ramsey, Bogleheads).

In a review scoring the 100 most-cited productivity hacks on usefulness and ease of use, timeboxing came out #1 (Zao-Sanders, HBR 2018). It works for a few compounding reasons. A time box is an "implementation intention" ("at 09:00 I will do X"); across 94 tests these if-then plans produced a medium-to-large lift in goal achievement (Gollwitzer & Sheeran, 2006, d ≈ 0.65). A fixed end time also caps Parkinson's Law, the tendency for work to expand to fill whatever time is available. And one block, one task means less "attention residue" bleeding from the previous task into the current one (Leroy, 2009). In Matteca, flip on Timeboxing in Today's Plan and the AI lays out back-to-back blocks with real start and end times, then you can push them to your calendar with reminders.

Zao-Sanders, HBR (2018); Gollwitzer & Sheeran (2006); Leroy (2009).