Why Preparation Time Does Not Double When Guest Does

When you cook for more people, you buy more rice. That part is easy: twice the serves, twice the rice. Time is harder. Chopping four onions is not twice as long as chopping two. A bigger pan of food is not twice as long on the stove.

The recipe box on this site scales ingredients by serves. It scales prep and cook with two different formulas. Both are guesses with known limits.

The ratio

Let r = current serves ÷ recipe serves.

For kande pohe, cooking for 4 means r = 4 ÷ 2 = 2. Cooking for 1 means r = 0.5.

Prep ≈ t × r^0.7

Prep is mostly hand work: rinse, chop, measure. You pay some cost once (find the board, wash the colander). After that, more food adds time, but not one-for-one. You get faster as you repeat the same cut.

The site uses:

prep = base prep × r^0.7

Why 0.7? Doubling (r = 2) gives 2^0.7 ≈ 1.62 , about 60% more prep, not 100%. That matches a common kitchen rule of thumb: doubling a batch adds roughly half again the chopping, not a full second batch from scratch. Meal-prep tools use the same idea when they say doubling portions adds about 60% more prep time, not 100%.¹

Breaks when: prep is mostly waiting (soaking beans, resting dough), or when you add a second cook and work in parallel.

Kande pohe — prep

Base: 10 min for 2. The solid line is what the site shows. The dashed line is naive multiply (10 × r).

• Site: 10 × r⁰·⁷
• Linear: 10 × r

At 4 serves the site shows 16 min prep; linear scaling would say 20. At 8 serves: 26 min vs 40.

Serves	r	Site prep	Linear prep
1	0.5	6 min	5 min
2	1	10 min	10 min
4	2	16 min	20 min
8	4	26 min	40 min

For one person the site is a little higher than pure half (6 vs 5): you still set up the board and find the turmeric. For a crowd it stays well under linear.

Cook ≈ t × ∛r

Cook time is about heat reaching the middle of the food. Heat moves from the outside in. If the dish grows in three dimensions (a thicker roast, a deeper pot) and keeps a similar shape, the distance to the centre grows like the cube root of volume. Volume here tracks serves, so I use:

cook = base cook × ∛r

Doubling (r = 2): ∛2 ≈ 1.26 , about 26% more time on the heat, not double. That “path length” form is the simple model used by several cooking-time scalers.²

Thickness and shape dominate clock time more than raw weight alone.

A stricter diffusion model says time often scales closer to weight^⅔, because diffusion time grows with thickness squared, and thickness grows like weight^⅓. That gives doubling ≈ +59% cook time, not +26%.

This site uses the milder cube-root form on purpose: a lower bound that avoids the common mistake of doubling heating time, without pretending we solved the full heat equation for every pan. For thin food (chapati, stir-fry in one layer), even cube-root is often too much; time barely changes.

Breaks when: the food is thin and flat; you split into many small pieces; the burner cannot hold temperature under a bigger load; or you change method (steam instead of fry).

Kande pohe — cook

Kande pohe is a wide pan of flakes, not a roast. Cube-root is a soft fit: more food may need a wider pan or a slightly longer steam, not a deeper thermal core. Still useful as “don’t double the 12 minutes.”

• Site: 12 × ∛r
• Linear: 12 × r

At 4 serves the site shows 15 min cook; linear would say 24. At 8 serves: 19 vs 48.

Serves	r	Site cook	Linear cook
1	0.5	10 min	6 min
2	1	12 min	12 min
4	2	15 min	24 min
8	4	19 min	48 min

Linear cook at 8 serves (48 min) would ruin pohe. The site’s 19 min is still only a guide: taste and steam until the flakes soften.

Both together

Total “active card time” on the recipe is prep + cook. For kande pohe at the written base that is 22 min. Here is how that total grows under the site’s formulas versus multiplying everything by r.

• Site (prep⁰·⁷ + cook∛)
• Linear (both × r)

The gap widens fast. At 8 serves: about 45 min on the site’s curves vs 88 min if you scale times like ingredients.

For kande pohe, that means four people are closer to “a bit more chopping and a slightly longer steam” than “double the morning.” Trust the pan over the timer when they disagree.

Footnotes

1. Batch-efficiency framing for meal prep (doubling ≈ ~60% more prep, not 100%): Bento Bunny — Meal Prep Time Estimator. ↩

2. Cube-root size factor for cook time (time × ∛size): CookCalcs — Cooking Time Scaler. ↩