Recipes
gelato base
fior di latte
white base

Gelato Base Recipe: One White Base for Every Flavor

Marco Freire — gelatiere & founder of Free Gelato Balancing App
Marco Freire
Gelatiere & founder
9 min read
Two small ceramic cups of white fior di latte gelato with clean quenelles on a marble counter
Two small ceramic cups of white fior di latte gelato with clean quenelles on a marble counter

One balanced white base, and every other flavour becomes an edit rather than a new recipe. This is a 1000 g fior di latte mix built to survive freezing without a machine, with the full arithmetic behind each number and the rules for turning it into nut, chocolate or fruit gelato.

Two small ceramic cups of white fior di latte gelato with clean quenelles on a marble counter A white base is the only recipe worth memorising. Everything else is an adjustment to it.

Weighed white gelato base ingredients in small ceramic dishes on a marble counter

What "One Base" Actually Means

A mother base is a mix balanced so tightly that adding a flavouring ingredient moves one number at a time, and you know in advance which one. Add pistachio paste and fat climbs. Add cocoa powder and solids climb while water falls. Add fruit purée and both sugar and water climb. If your starting point is already balanced, each of those is a single correction. If it is not, every flavour becomes a fresh problem.

This base is deliberately different from the machine version in fior di latte recipe. It is built for a static freeze, which means higher PAC and higher total solids, because there is no dasher shear to keep crystals small during freezing. Use it with the food processor method or the scrape-and-beat approach in how to make gelato without a machine. If you do have a mantecatore, drop the PAC back toward 250 and the solids toward 36% and it works there too.

Fior di latte hides nothing. There are no eggs, no vanilla and no cocoa to cover a mistake, so the balance carries the whole product. That is exactly why it makes a good mother base: if it tastes right on its own, it will taste right under a flavour.

The 1000 g White Base

Weigh everything. Never measure dairy by volume.

IngredientGramsRole
Whole milk (3.5% fat)575Water, milk fat, MSNF
Heavy cream (35% fat)165Fat for body and melt
Sucrose145Primary sugar, structure and sweetness
Dextrose (anhydrous)40Freezing point control, less sweet per gram
Skim milk powder46MSNF without added water
Maltodextrin DE1925Solids at PAC 0, no sweetness
Stabiliser blend (LBG and guar)4Water binding, slower melt
Total1000

That lands at 7.8% fat, 10.4% MSNF, 18.5% sugars, 39.5% total solids and PAC 278.

The maltodextrin is the piece most home recipes leave out and the reason this base holds up without a machine. It raises total solids by 2.5 percentage points while contributing nothing to sweetness and nothing to PAC, so it takes free water off the table without pushing the mix soft or sugary. Maltodextrin DE19 is the usual choice; inulin does the same job if you prefer it and adds a little fibre.

Saucepan of pale gelato base heating with a digital probe thermometer resting in it

The Numbers, and Why Each One Is There

Quick reference. Target 6 to 8% fat, 10 to 11% MSNF, 18 to 19% sugars, 38 to 42% total solids and PAC 270 to 290 for a no-machine white base served at -12 to -14 C.

Chart plotting the 1000 g white base against its no-machine target bands: fat 7.8%, MSNF 10.4%, sugars 18.5%, total solids 39.5% and PAC 278, every figure inside its target range Figure 1 - each balance figure for the 1000 g base against its target range.

Here is the arithmetic, using the standard composition figures: whole milk 3.5% fat and 8.7% solids-not-fat, cream 35% fat and 5.4% SNF, skim milk powder 97% solids, and lactose at 54.5% of MSNF.

FigureCalculationResultTarget
Fat(575 x 0.035) + (165 x 0.35)77.9 g, 7.8%6 to 8%
MSNF(575 x 0.087) + (165 x 0.054) + (46 x 0.97)103.6 g, 10.4%10 to 11%
Sugars145 + 40185 g, 18.5%18 to 19%
Total solidsfat + MSNF + sugars + maltodextrin + stabiliser395.5 g, 39.5%38 to 42%
PAC145 + (40 x 1.90) + (56.5 lactose x 1.00)278270 to 290

Three of those deserve a sentence each.

MSNF sits at 10.4% on purpose. Push past roughly 11% and the lactose load exceeds what the remaining water can hold in solution during storage, and lactose crystals give you the sandy texture that no amount of churning fixes. That ceiling is why the base uses maltodextrin rather than more skim milk powder to reach its solids target.

The sugar split is 145 sucrose to 40 dextrose, not 185 sucrose. Dextrose has a PAC of 190 against sucrose at 100, so it lowers the freezing point almost twice as hard per gram while tasting less sweet. That pairing buys the scooping softness a no-machine batch needs without pushing the mix into cloying territory. If the result tastes right but freezes hard, shift five grams from sucrose to dextrose and recheck in the PAC calculator.

Stabiliser at 0.4% is not optional here. In a batch freezer you can get away with none. In a static freeze the locust bean gum and guar blend is the only thing slowing crystal growth over six to ten hours. Blend the powder with a spoonful of the sucrose before it meets liquid or it will clump.

Process: Pasteurise, Age, Freeze

  1. Blend the dry. Combine the stabiliser with a small handful of the sucrose, then mix in the skim milk powder, maltodextrin and remaining sugars. This prevents the classic stabiliser fisheye.
  2. Warm the dairy. Bring milk and cream to about 40 C, then rain in the dry blend while whisking.
  3. Pasteurise. Hold 65 C for 30 minutes, or take it to 85 C briefly. The US frozen dessert standard (21 CFR 135.3) sets the equivalent floor at 155 F (68 C) for 30 minutes or 175 F (79 C) for 25 seconds. The lower, longer hold preserves more fresh-milk flavour; the higher, shorter one buys a longer mix shelf life.
  4. Chill fast. Below 4 C within 90 minutes, ideally over an ice bath.
  5. Age. 4 to 12 hours at 4 C. Maturazione is where the stabiliser fully hydrates and the fat partially crystallises, and it is the cheapest quality gain available to you.
  6. Freeze. Churn it, or freeze it in a shallow covered metal tray at 25 mm depth and process it frozen.
  7. Serve at -12 to -14 C. Expect roughly 10% overrun without a machine, against 20 to 35% with one.

Covered glass bowl of pale gelato base ageing cold on a marble shelf

Turning the Base Into Any Flavour

The rule is always the same: add the flavouring ingredient, see which number it moved, and correct that number by adjusting the ingredient that owns it. Nothing else changes.

Add per 1000 g of baseWhich number movesThe correction
80 to 100 g pure nut pasteFat climbs by 4 to 6 pointsCut cream by 100 to 130 g, replace with milk
60 to 80 g cocoa powder (22/24 fat)Solids climb, water falls, bitterness risesAdd 60 to 80 g milk, add 10 to 15 g sucrose
250 to 300 g fruit puréeSugars and water both climbReplace an equal weight of milk, recount fruit sugars in PAC
40 to 60 g couverture chocolateFat and sugar climb togetherCut cream by 60 g and sucrose by 25 g
Up to 2% alcohol by volumePAC climbs steeplyCut dextrose first, then sucrose
Variegato or inclusionsNothing in the balanceLayer in after freezing, cost only

A few specifics worth knowing. Hazelnut and pistachio pastes are roughly 55 to 60% fat, which is why the cream cut is so large; use 100% pure paste with no added sugar or oil, or the arithmetic above will not hold. Cocoa powder behaves as a dry solid with real water-binding capacity, so a chocolate version needs extra liquid or it turns pasty. Alcohol has an outsized effect on freezing point, which is why the cap is so low.

For fruit, understand that you are no longer making gelato with fruit in it; you are making a fruit gelato with a different target window. Count the sugar the fruit brings before you decide how much sucrose to keep.

Scaling, Storing and the Mistakes That Cost Most

Scaling is straightforward because the recipe is expressed per 1000 g: multiply every line by the same factor. For odd batch sizes use the recipe scaler rather than rounding by hand, because rounding the stabiliser is exactly the kind of small error that shows up as texture.

The mix keeps refrigerated at 4 C for up to 72 hours after pasteurisation, and it improves for the first 12 of those. Finished gelato is a different story: without a machine, plan to serve within 24 to 48 hours before recrystallisation coarsens it.

Three mistakes account for most failed batches. Measuring dairy by volume instead of weight puts the fat figure out by a full point before you start. Skipping the ageing rest leaves the stabiliser only partly hydrated, which shows up as a thin, fast-melting body. And adding flavour paste without touching the cream pushes fat past 12%, where the mix stops tasting clean and starts coating the palate. If your finished batch is disappointing, work through why is my gelato not creamy with these numbers in hand.

Two ceramic cups of white fior di latte gelato with clean quenelles on a marble counter

Edit this base live. Open the Free Gelato Balancing App and watch PAC, POD, total solids, MSNF and fat update as you change any ingredient weight. Free, web-based, no install.

gelato base
fior di latte
white base
gelato recipe

Frequently asked questions

Common questions about recipes.

Continue reading

View all

You read the method. The book has the batch.

Twenty-four recipes balanced to the gram, five ways to freeze them without a machine, and the numbers behind every one. Instant PDF, yours to keep.

See the book

One payment. No subscription.

Balancing your own recipe instead? Open the free balancer