If you make filled chocolates, one question follows every ganache recipe you write: how long will this actually last? Not in a sealed lab. In a box, in a shop, at room temperature. The answer depends almost entirely on a single number called water activity (written as aw).
What is water activity and why does it matter?
Water activity measures how much of the water in a food is available to support microbial growth — as opposed to water that is bound up in sugar, proteins or fats and cannot move freely. It runs on a scale from 0 (bone dry) to 1.0 (pure water). Ganaches typically fall in the range of 0.60 to 0.88 — and where in that range your recipe sits makes an enormous difference to shelf life.
- Below 0.65: most bacteria, yeasts and moulds cannot grow. Very long shelf life.
- 0.65–0.75: stable against most spoilage, but some xerophilic moulds can operate here.
- 0.75–0.85: short to medium shelf life — typically two to four weeks for ganache.
- Above 0.85: bacteria become active. The ganache should be treated as a short-life product, usually one to two weeks at most.
What affects water activity in a ganache?
The main levers are the ingredients that either contribute free water or bind it:
- Cream and milk are the biggest source of free water. The higher the proportion of cream in a ganache, the higher the water activity.
- Sugar binds water through its strong affinity for water molecules. Invert sugar and glucose syrups are particularly effective at this — they also prevent crystallisation, which is why chocolatiers use them.
- Chocolate contributes cocoa solids and fat, both of which reduce water activity. Higher cocoa-percentage chocolate tends to lower it more.
- Butter is mostly fat and lowers water activity, but it adds relatively little compared to sugar.
- Alcohol (brandy, rum, liqueurs) reduces water activity and has direct antimicrobial properties of its own — useful but easy to over-rely on.
- Fruit purées are high in free water and often raise water activity significantly, even if they feel “dry”.
How the maths works (in simple terms)
True water activity calculation uses something called Raoult’s law and, for real food systems, more complex models like Ross or Norrish. The short version: each ingredient contributes to the total water activity based on how much free water it contains and how well it binds water. The ratios matter more than absolute amounts — a recipe that is 30% cream has a higher water activity than one that is 10% cream, regardless of batch size.
Chocolatiers who formulate commercially often use purpose-built tools to do this maths — the ChocoX Shelf-Life Predictor uses a machine-learning model trained on measured-aw recipes to give a fast practical estimate without needing to understand the underlying equations.
The lab vs. the estimate
A laboratory water-activity meter (an aw meter) is the only way to measure water activity directly. These instruments typically cost £500–£3,000, which puts them out of reach for most small producers. Commercial testing services exist — they will test a sample for £40–£100 — and for anything you sell formally, a real test is the right answer.
But for recipe development — deciding whether to tweak the cream ratio before you commit to a batch, or understanding why a new recipe seems to go off faster than your usual one — a calculated estimate is very useful.
The ChocoX Shelf-Life Predictor was tested against more than 250 recipes in Callebaut’s Chocolatier’s Kitchen. It agreed with the book or was more cautious on about 93% of them, and flagged 52 of the 55 genuinely short-life recipes.
Practical steps for a chocolatier
- Write out your recipe by weight in grams — every ingredient.
- Use an estimator (or calculate manually) to get a predicted water activity range.
- If the estimate is above 0.80, treat this as a short-life product and label accordingly.
- If you want to extend shelf life: add more invert sugar or glucose syrup, reduce cream proportion, or increase chocolate percentage — then re-estimate.
- For anything you sell commercially, confirm with a real keeping trial or a lab test.