How We Source Nutritional Data for Our Diets
Learn about how we source nutritional data for our diets.
Our Approach to Nutritional Data
When developing diets at Ayla's Kitchen, the quality of the nutritional data we use is fundamental. It is not only a question of what nutrients an ingredient contains, but also how those values were derived, which assumptions they are based on, and what limitations exist within the data itself. Our aim is to work with the most robust and reliable data available, while at the same time being open about uncertainty where it exists—both in the data and in the methods used to interpret it.
USDA as Our Primary Data Source
The majority of the nutritional data we use is sourced from USDA FoodData Central, one of the most comprehensive and detailed food composition databases in the world. It covers a wide range of raw materials and ingredients, providing data on both macro- and micronutrients, often across multiple forms of the same ingredient—such as raw, cooked, processed, unsalted, or prepared using different heat treatments. A key reason for choosing USDA FoodData Central as our primary source is that it is built on standardised laboratory analyses and well-documented calculation methods. In addition, the database is openly accessible to the public. This means the data is fully transparent and verifiable—not only for us, but also for other professionals and interested users. In a field that can often feel opaque, this openness is a deliberate and important choice for us.
When Data Is Incomplete
Even in a database as extensive as USDA FoodData Central, limitations exist. Certain micronutrients—such as choline, iodine, biotin and chloride—are not consistently reported for all ingredients. These nutrients are relevant in canine nutrition and are also addressed in nutritional guidelines such as those issued by FEDIAF. When such values are missing, we supplement our data using other authoritative food composition databases and scientific sources. The national databases we rely on have been selected based on methodological quality, scope and credibility, and include: - Norwegian Food Composition Table - Swiss Food Composition Database - DTU Food Composition Database (FRIDA, Denmark) - New Zealand Food Composition Data - Canadian Nutrient File - Australian Food Composition Database Where appropriate, we also draw on peer-reviewed scientific literature to support informed and responsible nutritional assessments.
Market-specific Products and Conservative Assumptions
Some of the ingredients we use are commercial or market-specific products for which complete nutritional profiles are not publicly available. In such cases, nutrition labels or local data sources may only provide macronutrients and a limited selection of micronutrients. When directly comparable data cannot be found, we apply conservative assumptions and carefully considered conversions, based on the closest relevant data and established scientific methodology. For example, comprehensive nutrient data may exist for raw chicken liver, while data for pan-fried liver is unavailable. In these situations, we estimate nutrient content using documented retention and loss factors from the scientific literature, rather than assuming unchanged values.
Preparation Methods and Practical Variation
Food preparation has a significant impact on nutrient composition. The databases we use typically provide multiple preparation options for the same ingredient—such as boiled, pan-fried, braised or steamed—and we actively use these distinctions in our calculations. At the same time, it is important to recognise that database terms like “boiled” or “stewed” cannot fully capture how food is prepared in real kitchens. Small differences in temperature, cooking time, liquid volume, and whether cooking liquid is retained or discarded can all influence nutrient levels, particularly for water-soluble nutrients. As a result, some variation between laboratory values and practical preparation is inevitable.
Natural Variation in Fresh Ingredients
There is also inherent natural variation within raw ingredients themselves. No two avocados, cuts of meat or vegetables are ever identical. Nutrient composition is influenced by factors such as geographical origin, soil mineral content, climate, and production conditions. Although laboratory analyses typically rely on multiple samples to establish average values, nutrient content can still vary between batches and over time. This is also reflected in the small differences often seen between national food composition databases for the same ingredient. For this reason, we treat nutritional values as scientifically grounded averages, rather than exact values for every individual raw material. This is a well-established and accepted principle in nutritional science, and one we keep in mind when interpreting and applying available data.
Our Perspective
We recognise that nutritional data always represents an informed scientific estimate, not an absolute truth. Natural variation in ingredients, differences in preparation, and limitations in available data mean that some degree of uncertainty will always exist. Even with careful sourcing, cross-referencing between databases, and the use of established methodologies, errors can occur. In addition, new products, ingredient variations and market-specific raw materials continuously emerge—sometimes without complete nutritional data being immediately available. For this reason, we actively welcome dialogue with our users and professional peers. If something appears unclear, inconsistent, or open to improvement, we encourage engagement and feedback. We see this exchange not as a correction mechanism, but as an important part of maintaining quality over time—and as a way to continuously refine and strengthen the data and assumptions behind our diets.
Database-sources
USDA FoodData Central: https://fdc.nal.usda.gov/food-search?type=Foundation Norwegian Food Composition Table: https://www.matvaretabellen.no/ Swiss Food Composition Database: https://naehrwertdaten.ch/en/ DTU Food Composition Database (FRIDA, Denmark): https://frida.fooddata.dk/?lang=en New Zealand Food Composition Data: https://www.foodcomposition.co.nz/ Canadian Nutrient File: https://food-nutrition.canada.ca/cnf-fce/newSearch Australian Food Composition Database: https://afcd.foodstandards.gov.au/default.aspx