snapkin
The camera · Decision 3 of 14
Why we track added sugar, and what we mean by it
Snapkin says “added sugar” everywhere, because that is the phrase people use. Underneath, it uses the World Health Organization's definition of free sugars. That covers more.
Last updated .
- What the app does
- Counts WHO free sugars, and counts fruit purées too. The daily limit is roughly 10% of your calorie target. No streak on it, no reward for staying under.
- How sure we are Thin , One small study, or an engineering choice.
- Less than anything else the app shows. It is a rough budget, not a measurement. That is why it has no confidence badge.
The evidence for it, and there is little of it
This is the weakest evidence on any of these pages, and the page says so throughout. Here is what there is.
What “added sugar” means here
The definition underneath is the World Health Organization's free sugars: sugar added by a manufacturer, a cook or the person eating, plus honey, syrups, and fruit juice and juice concentrates. It does not include the sugar in whole fruit, in vegetables, or in plain milk and yoghurt. The app also counts fruit purées and smoothies. That follows the wider 2018 UK definition, which looks at how much the fruit was processed. The WHO’s wording does not list all of these processed-fruit cases. Plain stewed or canned fruit that still has pieces in it is not automatically free sugar. Any sweetener added to it still counts.
The “Added Sugars” line on a US nutrition label (21 CFR 101.9) is not the same thing. Total sugars, the figure on a European label, is a third thing again. The three definitions differ on juice, and on the sugar that is naturally in whole foods.
| Counted as sugar? | Snapkin (WHO free sugars) |
US label “Added Sugars” |
EU label “of which sugars” |
|---|---|---|---|
| Sugar stirred in, or added by a manufacturer | Yes | Yes | Yes |
| Honey, syrups, molasses | Yes | Yes | Yes |
| 100% fruit juice, including correctly reconstituted juice | Yes | No | Yes |
| Concentrated juice used as a sweetener | Yes | Yes* | Yes |
| Sugar in whole fruit and vegetables | No | No | Yes |
| Lactose in plain milk and yoghurt | No | No | Yes |
*US rules contain specific concentrate exceptions. See FDA guidance and the regulation linked above.
The WHO guideline recommends keeping free sugars below 10% of energy, and suggests going below 5% if possible. Snapkin takes 10% of your target calories and turns it into grams at 4 kcal/g. Then it rounds to the nearest 5 g. So the budget you see can be a little above or below 10%. It is a rough budget for tracking, not an exact WHO limit.
How the number is worked out
Snapkin gives the analyser this definition and a step-by-step guide for estimating. The definition decides which sugars count. The guide decides what to do when a label or recipe is incomplete. Knowing the definition does not tell it the ingredients or the portion size of a photographed meal.
Free sugars are not printed on any European label, so they have to be estimated. Experts have published methods for estimating added and free sugars from food tables and ingredient lists. The instructions we give the analyser borrow their order of steps. But an AI following instructions is not the tested expert method. Louie et al. (2015) set out ten steps, starting from hard evidence and ending with guesses by food category. Kibblewhite et al. (2017) restated them for WHO free sugars. Scapin et al. (2021) adapted them for a country whose labels show neither total nor added sugars.
Scapin checked the method against the values on US labels for 930 products. The agreement score (ICC) was 0.98. Davies et al., 2022 tested Louie's method on their own. They found R² 0.97 and a mean absolute error of 1.26 g per 100 g. Davies used a made-up test set of 500 Australian products to compare the manual method with a machine-learning method. Both tests are about packaged food information, not photographs.
The repeat study asks a different question. Two people followed the same written steps for the same 5,740 foods. Their answers differed with a standard deviation of 4.6 g per 100 g. You cannot divide that number by another study’s mean error and call the result “so many times less accurate”. The measures, the samples and the comparisons are all different.
The evidence against, and there is a lot of it
The weight effect is mostly about calories
Te Morenga et al. (BMJ 2013): eating less sugar changed weight by −0.80 kg (95% CI −1.21 to −0.39). Eating more changed it by +0.75 kg (0.30 to 1.19). But there was no significant weight difference when sugar replaced other carbohydrates at the same calories. This is a finding about body weight. It is not proof that sugar has no other health effects. The WHO’s guideline also looks at tooth decay.
Estimating sugar from a photograph is the weakest thing this app does
One peer-reviewed photo study reports total sugar: O'Hara et al. (2025), 114 photographs of weighed meals, judged by ChatGPT-4. It measured total sugar, not free sugar. We have no test showing that Snapkin can tell the two apart from photographs. Hidden ingredients make that a very uncertain estimate.
“Carbs that turn into sugar” is a different thing, and we do not track it
Carbohydrate your body can digest, including starch, can raise your blood sugar. Glycaemic load combines a food’s glycaemic index with how much carbohydrate it has. It is not the same thing as free sugar. Zeevi et al. (Cell 2015) gave 800 people continuous glucose monitors across 46,898 meals. Different people responded very differently to the same meal. Published glycaemic index and load tables exist, but they do not predict every person’s response. Snapkin does not estimate your personal blood sugar response.
So the sugar limit is a rough budget, not a measurement. It carries no confidence badge. The eating disorders page has the other half of this decision: there is no streak on the sugar limit, and no celebration for staying under it.