I want to be upfront about what this post is and is not.
It is not a tutorial. There is no code to copy. It is a field note about the one thing sitting underneath every feature any of us has ever shipped: discipline. I just finished 75 Hard, a 75 day program with five daily non-negotiables and one brutal rule, and it changed how I think about consistency at work more than any productivity system I have tried. So I sat down and mapped what those 75 days actually taught me about engineering.
If you build things for a living, stay with me. The parallels are closer than they look.
What 75 Hard actually is
75 Hard is not a fitness challenge. Its creator, Andy Frisella, calls it a mental toughness program, and the design makes that clear. Five tasks, every single day, for 75 days:
| # | Daily task | The catch |
|---|---|---|
| 1 | Follow a structured diet | Zero cheat meals, zero alcohol, no deviations |
| 2 | Two 45 minute workouts | One must be outdoors, and they must be 3 hours apart |
| 3 | Drink one gallon of water | Roughly 3.8 litres, every day |
| 4 | Read 10 pages of non fiction | Physical pages only, audiobooks do not count |
| 5 | Take a progress picture | Every day, no exceptions |
Then the rule that makes the whole thing hard: miss any task, on any day, and you start over at Day 1. Not the task. The whole program. A slip on Day 60 sends you back to zero.
Why almost nobody finishes
There is no official completion statistic for 75 Hard, and the "less than 1 percent finish" number that gets passed around online is not backed by hard data (treat it as folklore, not fact). But you do not need a statistic once you look at the structure:
| Over 75 days, finishing means | Total |
|---|---|
| Workouts completed | 150 (2 a day, no rest days) |
| Outdoor workouts, rain or shine | 75 |
| Gallons of water | 75 |
| Pages read | 750 minimum |
| Daily task check ins that must all hit | 375 |
| Misses allowed before you restart | 0 |
No rest day is the part people underestimate. One workout a day is a habit. Two workouts a day, three hours apart, with one of them outdoors regardless of weather, is a logistics problem you have to re solve every single day around work and life. And the restart rule means the difficulty is not linear. The cost of a mistake grows the further you get, because you have more to lose.
The science, and where 75 Hard deliberately disagrees with it
Here is the part that fascinated me as someone who likes systems.
The best real study on how habits form is Lally et al., 2010 (University College London, published in the European Journal of Social Psychology). They tracked 96 people forming a new habit and found it took a median of 66 days for a behaviour to become automatic, with a huge range (18 to 254 days depending on the person and the behaviour). The popular "21 days to a habit" line is a myth.
75 Hard runs for 75 days. That is not a random number. It is comfortably past that 66 day average, which lines up with exactly what I felt: the first week was pure willpower, and somewhere in the back half the tasks stopped being a fight and started being a routine.
Week 1 is the wall. By the end, the hard part runs on autopilot. Habit timeline based on Lally et al., 2010.
But here is the twist. That same study found something 75 Hard flatly ignores: missing one opportunity did not materially affect habit formation. Science says one slip does not break you. 75 Hard says one slip resets everything.
So is the restart rule irrational? I do not think so, and this is the insight I took to work. The restart rule is not about biology. It is a forcing function that kills the negotiation. It removes the "just this once" conversation entirely, because the downside is total. When there is no room to compromise, you stop spending energy deciding whether to. If you have ever watched "we will fix it later" quietly become permanent tech debt, you already understand why removing the negotiation is powerful.
The system that got me through
Motivation did not carry me through 75 days. A system did. I did the boring engineering thing: I removed willpower from the loop wherever I could.
- I started on a Friday, on purpose. My family was flying in the next morning and I already knew the weekend would be a tour of the best food in Delhi, a lot of it deep fried. Every sensible voice said start Monday. I started Friday specifically because if the discipline could survive its hardest possible test on day two, the other 73 days would feel easy. It worked. Surviving that weekend is what made the rest a formality.
- I made the gallon impossible to forget. I bought a one gallon bottle with time markers printed down the side. No tracking app, no math, no decisions. The bottle told me if I was behind. That is just a progress bar with a deadline.
- I fixed the variables. I ate the same clean meal on repeat so food was never a decision: 250g chicken cooked in chicken broth instead of oil, 150g broccoli, 100g bok choy, and four egg whites plus one whole egg scrambled with Greek yogurt, all cooked on water, zero oil. Boring on purpose. A decision you make once is not a decision you fight daily.
- I offloaded the prep. I have house help, and I will be honest about that rather than pretend I did this on hard mode. The meals were prepped while I was at the gym, so I came back to clean food ready to eat. Not everyone has that, and it genuinely lowered the difficulty. What is portable is the principle underneath it: take the highest friction, most willpower draining step and automate it out of your day. For most developers that step is not lunch, it is a script, a template, or a CI check.
How 75 days of this maps to shipping software
This is the part I did not expect. The habits transferred.
| 75 Hard | The engineering equivalent |
|---|---|
| Miss once, restart from Day 1 | Keep main green. One red test stops the line, no "merge it anyway" |
| Two workouts, no rest days | Consistent cadence beats heroic sprints. Ship small, ship daily |
| One workout must be outdoors | Step away to solve it. The bug gets fixed on the walk, not at the desk |
| A gallon of water, 10 pages, every day | Tiny non negotiables compound: one test, one review, one paper a day |
| Daily progress picture | You cannot improve what you do not measure. Instrument everything |
| The diet, the hardest rule | Saying no is a skill. Scope discipline is how projects stay shippable |
The throughline is the same lesson in both places: consistency is a system you build, not a mood you wait for. The developers I admire are not the ones with the most motivation. They are the ones who made the right thing automatic so it happens on the bad days too.
The honest caveats
75 Hard is rigid by design, and rigid is not the same as healthy for everyone. Two workouts a day with no rest days and a gallon of water daily is genuinely a lot, and it is worth talking to a doctor before starting rather than after. This is a reflection on discipline, not medical or fitness advice. And as I said, the house help that let me offload meal prep is a real advantage I would rather name than hide.
None of that changes the core takeaway, which is the most useful thing I have relearned in a while: the work does not require you to feel like it. It requires you to build a system that shows up whether you feel like it or not.
I finished on Day 75 with zero misses. The gym results were the least interesting part.
If you have done 75 Hard or anything like it, I am curious: what broke first, the diet or the no rest day? And for the engineers here, what is the one "keep main green" style rule that changed how your team ships? Tell me below.
Top comments (0)