Build Your Own Nutrition Plan

For three years I wrote nutrition plans for other people. The plans were the deliverable, but they were never the thing that worked. The system around them was. A plan is a document. A system is what makes the document survive contact with a Tuesday.

This is that system, written out so you can run it on yourself. It is the same sequence I used with about forty people, minus me.

I should say the obvious thing first: this is not medical advice and I am not your clinician. If you are managing a diagnosed condition, are pregnant, are underweight, have a history of disordered eating, or take medication that interacts with food, the plan you need is one built with a professional who knows your history. What follows is for an ordinary healthy adult who wants to eat better and keeps failing to.

Where to start, depending on who you are

This works at any age. The single place age changes the answer is protein, and step 3 says where.

Step 1: The baseline week

Before you change anything, find out what you currently do.

For one week, write down everything you eat. Everything. The handful while walking past the kitchen, the second helping, the thing after dinner that happened after dinner was mentally over. Include timings and roughly how hungry you were.

This is the step everybody wants to skip and the step that most often makes the rest unnecessary. When I did it honestly on myself, I discovered that my planned meals already added up to a fat-loss day and the entire problem was food I was not counting as eating. No new plan would have fixed that. Only the log showed it.

Do not change your eating during this week. A baseline you have improved is not a baseline.

Step 2: The self-interview

I always started a client with a phone call rather than a form, and the questions I actually cared about were not about food. Ask yourself these, and write the answers down rather than thinking them.

The routine

• What time do you wake, eat, train, and sleep? Which of those is fixed and which drifts?
• Where does your food come from? Do you cook, does someone else, do you order?
• How often do you eat out, and is it choice or logistics?

The body

• Hunger: when does it actually hit? Is it clock-driven or event-driven?
• Digestion: anything that reliably disagrees with you?
• Any medication, condition or restriction that a plan must respect?

The part that matters most

• What does food mean in your life? Are you eating for hunger, for comfort, for stress, out of habit, for company?
• Do you eat fast or slowly?
• What are the foods you will not give up? Name them specifically.
• What is your stress like, and does it change how you eat?
• What was your favourite physical activity as a child?

That last one always surprised people. It is not idle. The quickest route back into movement is almost never the gym; it is something you already loved before exercise became a duty.

Why the interview beats the form. A form tells you what someone eats. A conversation tells you why, what they have already failed at, and which parts of a plan they are quietly going to refuse. You are looking for three things about yourself: your patterns, your preferences, and your mental blocks. The third one is where plans go to die, and it never appears in a food log.

Step 3: Set the targets

Only four numbers matter, and you can set all of them in ten minutes.

Protein. Researched · ISSN Position Stand, Protein and Exercise For an exercising adult, 1.4–2.0 g per kg of bodyweight per day, per the International Society of Sports Nutrition’s position stand, toward the higher end if you are eating in a deficit and want to keep muscle. Split it across meals at roughly 20–40 g each rather than loading one large dinner.

If you are over about sixty, or building this for someone who is: ageing muscle responds less strongly to the same dose of protein, so the usual adult minimum is widely argued to be too low. Commonly recommended instead is 1.0–1.2 g/kg/day at minimum, and, importantly, larger single servings, because a small dose may not clear the threshold to trigger a response at all. More protein, in bigger portions, fewer times. This is the most common gap I see in plans built for older parents.

Energy. Reasoned, not tested If fat loss is the goal, aim for a rate of roughly 0.5 kg per week. Slower keeps muscle and is far more likely to survive; faster is where the rebound lives. ⚠ This is a practitioner convention rather than a number derived from a specific trial. It is what I used with clients and what most coaches use, and I would rather tell you that than dress it up with a citation it does not have.

Sodium. Lower than you think, and not by touching the salt shaker but by dealing with pickle, papad, namkeen, packaged masala, sauces and restaurant food.

Plate shape. Half vegetables, a quarter protein, a quarter whole grains. This is a proportion rule, not a weighing exercise. Nobody sustains weighing.

Step 4: Write the plan, with options

Lived Here is the mechanism that mattered more than anything else I did as a coach.

Never write one meal per slot. Write two or three.

A plan with a single option is a prescription, and a prescription fails the first time the ingredient is missing, the day runs late, or you simply do not want it. That failure does not stay local. It becomes “I am off the plan,” and by Thursday the whole thing is abandoned. Options mean a bad day is a substitution, not a collapse.

Some things that worked:

• Timed slots, two or three options each. Breakfast, lunch, snack, dinner.
• Every slot has its protein anchor named. Decide protein first, build the rest around it.
• Dessert is on the plan. A few times a week, scheduled, not negotiated in the moment. Every attempt I made to remove it failed. Budgeting it removes the guilt, and the guilt was doing more damage than the sugar.
• A carry kit. Roasted chana, dates, a bar, permanently in your bag. The junk default is a logistics problem, not a moral one.
• Portions in units you already use. A chapatti is seven inches. Two katoris make a bowl. Four handfuls of vegetables a day, plus a fruit.
• Print it and put it on the fridge. This sounds quaint. It outperformed every app notification I have ever seen.

Step 5: The weekly check-in

Every Friday, my clients got the same thing: a short form, a nudge to update their food diary, and an open offer to talk at the weekend. The stated purpose, in my own words at the time, was “to have a clear picture of what we did during the week so we can become more aware of our habits and patterns.”

Awareness first. Adjustment second. Most weeks required no change to the plan at all. The value was in looking.

Doing this alone, weekly, ask yourself:

• What actually happened this week versus what was planned?
• Where exactly did it come apart? Which day, which meal, what was going on?
• What is one thing to change for next week? One. Not five.
• What worked that I should protect?

This step is the system. Steps 1 through 4 produce a document. Step 5 is what turns it into a practice, and it is the step people drop first because it feels like admin rather than progress. When my clients renewed, it was almost always because of the Fridays.

Step 6: The mess-up protocol

Write this down before you need it, because you will not write it well in the moment.

The version I gave clients, in the exact words I used: “Move on. Shit happens. We aren’t perfect.”

And the rule underneath it, which is the actionable part: always train the day after a mess-up. Not as penance, but because it converts guilt into a scheduled action, and guilt with nowhere to go is what turns one bad meal into a bad fortnight.

The single most useful thing I ever told anybody about eating is that missing once is close to irrelevant. Do not let a bad Tuesday cost you Wednesday.

Researched · Lally et al., European Journal of Social Psychology, 2010 That first half is not folklore. A study tracking 96 people forming a daily habit over twelve weeks found that missing a single opportunity did not materially affect habit formation; automaticity dipped by a fraction of a point and recovered. The same study found habits took a median of about 66 days to become automatic, with an enormous range of 18 to 254, which is its own useful antidote to the twenty-one-day myth. ⚠ The “never miss twice” half is a rule of thumb, not a finding. It comes from popular habit writing rather than from that study, and I repeat it because it works as a decision rule, not because anyone measured it.

Why plans fail: the coach’s view

Lived Having watched about forty people run this, the failures were remarkably consistent, and almost none of them were about nutrition:

• The plan was optimal rather than doable. The best plan you will not follow loses to a decent one you will, every time.
• It was built for a life the person did not have. Somebody else’s schedule, somebody else’s kitchen, somebody else’s budget.
• There was no fallback. One missing ingredient, and the day was declared a write-off.
• It fought something the person loved. Anything at war with pleasure loses eventually.
• Nobody was looking. Not judging, looking. The weekly review is the cheapest accountability there is, and doing it alone is harder than people expect, which is most of what a coach was actually for.

The kit

Everything above, as a file you can hand to whatever AI assistant you already use so it runs the interview with you and drafts the plan:

→ the nutrition plan kit

It carries the self-interview, the target maths, the plan format with options, the weekly check-in, and the mess-up protocol. Plain text, nothing to sign up for, nothing sent anywhere. You download it and it runs wherever you already work. It will ask you about medication and conditions only to tell you to take those to a doctor, not to plan around them. That boundary is deliberate and it is not one an assistant should be crossing.

The evidence behind this page

This page is mostly method, a workflow I ran with about forty people, rather than science. That distinction matters, so here it is explicitly.

⚠ The honest framing: the targets are evidenced, the workflow is craft. If someone tells you a coaching system is “science-backed”, ask which half they mean.

Cheat sheet for building the plan (📊 = evidence · 🔧 = my own method)

• 🔧 Log one honest week before changing anything. Do not improve the baseline.
• Interview yourself, including the questions that are not about food.
• 📊 Four numbers: protein (1.4–2.0 g/kg, more and larger doses past sixty), rate (~0.5 kg/week), sodium (low, at the source), plate (half veg, quarter protein, quarter grain).
• 🔧 Two or three options per slot. Never one.
• Schedule the dessert. Carry the snack. Print the plan.
• Friday: review, change one thing, protect what worked.
• 📊 Missing once barely matters (Lally 2010). 🔧 “Never miss twice” is a rule of thumb, not a finding.

Message me

If you run this and get stuck, especially at step 2, where the honest answers are harder than they look, message me. I am not coaching and I am not selling anything; the practice closed years ago. But I spent three years watching where people got stuck, and it is nearly always the same three places.

Email: abhishek@terrak.ai LinkedIn: Abhishek Chaudhry

What’s changed

• 31 August 2026. Added an evidence section separating what is actually evidenced (the targets) from what is my own craft (the workflow). Sourced the habit claim to Lally et al. 2010, and flagged that the “never miss twice” half of it is a rule of thumb rather than a research finding.
• 30 August 2026. Added provenance labels naming the source behind each claim.
• 30 August 2026. First published.