Procaaf
Practical·6 min read··By The Procaaf Team

How to Count Macros When Eating Out

Restaurant meals without a label: how to estimate calories and protein within 10%, which dishes are predictable, and where the hidden fat sits.

A restaurant table with plated dishes.
A restaurant table with plated dishes.

You cannot weigh a restaurant meal, but you can estimate it to within about 10% — which is close enough, because label tolerances are around 20% anyway. The method is to price the components separately rather than search for the dish name.

A grilled chicken bowl is not a mystery: it is a palm of chicken, a cup and a half of rice, some vegetables, and two to three tablespoons of oil or sauce you did not see go in.

Estimate by component, not by dish

Protein: a restaurant portion of meat or fish is typically 150–200 g cooked, which is a palm and a half. Grains: a standard side of rice or pasta is 200–250 g cooked. Vegetables: negligible calories unless glossy with oil.

Then add the invisible fat. Assume 15–30 g of cooking oil or butter for any pan-cooked or restaurant-sauced dish. That single assumption fixes most under-counting.

Predictable dishes vs. wildcards

Predictable: grilled meats, sushi, tandoori dishes, salads with dressing on the side, burrito bowls, roast dinners. What you see is roughly what you get.

Wildcards: creamy curries, stir-fries, fried rice, anything battered, restaurant pasta and most desserts. These vary by hundreds of calories between kitchens. If you order them, log a generous estimate and move on rather than pretending precision.

The one-meal-a-week rule

A single 800-calorie overshoot across a week is about a 115-calorie daily error — nothing. Four of them is a stalled diet. So estimate honestly, keep the rest of the day protein-heavy, and do not compensate by skipping meals beforehand; that usually leads to eating more, not less.

Build the estimate in the Procaaf scratchpad before you order if you can. Seeing '2 tbsp oil = 240 kcal' on the screen makes the sauce question answer itself.

Frequently asked questions

How accurate can restaurant macro estimates be?
Within about 10–15% if you estimate by component and assume added cooking fat. That is comparable to the tolerance allowed on packaged food labels.
Should I stop eating out while tracking?
No. Sustainability matters more than precision. Estimate generously and keep the frequency reasonable.
What is the safest thing to order?
Grilled protein with a starch side and vegetables, sauce on the side. It is the easiest meal to estimate and usually the highest in protein.
Keep reading

Continue reading