User guide
The database on Food Losses and Waste is a continuing endeavour by FAO under Sustainable Development Goal (SDG) 12.3 to share available knowledge and help focus efforts where they can have the most impact in order to meet the SDG target. This is a living dataset, based on an extensive and continuing literature review, so always check back for additions and improvements. We know it will be a collaborative effort to improve the data – so if you see an area where you have data, or would have a documentary source to be added, please send it to us!
The database is open for users to interact with the data. Want to know where tomatoes are lost? How about in your region? Here you can find all the options for refining your queries to the dataset.
The database is made of two main parts:
- a selection panel on the left which allows you to refine the search on a number of filters, and
- the data visualisations at the centre of the page where you can choose how to visualise the data by clicking on the tabs above the graph.
1. The selection panel - Filtering options
Several options are available to users in order to refine the information and zoom into the results that are key to their needs. Filters cover time, areas (3 levels), commodities (2 levels), stages and methods. Filters can be used independently of each other in most cases, so that a user could look for losses across all commodities for a geographic aggregation or vice versa – specific commodities or commodity groups across all geographies.
Year range
Select the period of interest. This is the year that the study was undertaken. Years range from 1965 to present.
Area aggregation
Select the broad geographic area of interest. Geographic aggregates follow the M49 geographic standard of the UN Statistics Division. Several alternative geographical aggregates such as Country, SDG regions, or the World Bank Income Classifications are also provided.
Area aggregates
Select the sub-groups within the Aggregation heading previously filtered.
Country
Individual Countries can be selected within the sub-groups in the Area aggregates. The list of countries follows the M49 standard. For example, under the ‘SDG Region’ (Area aggregation), and the ‘Latin America and the Caribbean’ (SDG=M49) (Area aggregates), only the countries in this region can be selected with the Country filter (Aruba, Anguilla, Argentina, etc.). If the data exists for the selected country(ies), the data points will be filtered and presented in the graphs.
Food groups
The data can be filtered for six food groups. The options in this filter are:
- Cereals & pulses
- Fruits & vegetables
- Roots, tubers & oil crops
- Animal products
- Fish products (to be added) and
- Other (stimulants, spices)
These are the same commodity groups used to compile the SDG indicator 12.3.1 (a) Food Loss Index and that countries will use to select their top ten commodities. The selection in this category restricts the choice of food commodities in the Commodity (CPC 2.1) filter, much like the geographic aggregation.
Select to include groups and broadly defined products
This option when selected will display the average loss data by groups of products e.g. cereals, pulses, fruits etc.
Commodity (CPC 2.1)
Individual commodities can be selected with this filter within the sub-group Basket Items. The list of items follows the international Central Product Classification version 2.1 (CPC 2.1) used in FAOSTAT. This is the international standard for definitions, concepts and classification of commodities and food products.
Value chain stage(s)
In an effort to improve the knowledge on where losses and waste are occurring within the commodity supply chains, the data points were tagged with the stage of the value chain that was described in the source study. They follow the main stages considered for the SDG indicator (point of production (farm, slaughter, landing); storage; transport; wholesale; processing); however, for some countries and commodities additional stages or activities have been included as they are critical loss points that were the objective of the study (click here for detailed description).
Method of data collection
Options for this filter categorize the main loss measurement/estimation methods found in the literature or point at the main data sources. Knowing the method used in each study helps in understanding its data scope or representativeness (click here for detailed description).
2. Data visualisations
Plot of loss percentages
It plots all the filtered data points. When opening the database and before selecting, the graph shows the content of the whole database as displayed below.
Each point represents a specific research/study. Loss percentages are plotted by year. A colour code indicates the different Methods of Data Collection.
When pointing the cursor on a point, a metadata box appears reporting the related year, actual loss percentage, commodity, country, method of data collection, data source, reference and link to the original source.
Heatmap of available data
It summarises the extent of available data by geographic area (region or country depending on the filters) and items (commodity groups or individual commodities depending on the filters) for the whole food chain.The color changes with the amount of available information. The graph therefore illustrates data scarcity or abundance. A white space signifies no data is available.
Boxplot by stage
The boxplot is organized by stage of the supply chain. It is a breakdown of our data set that can be used to identify the stage at which losses occur the most and therefore target intervention strategies.
In this database, it compares the levels of food loss at the different stages in the value chain. At every stage in the supply chain, it has a box to show the range where 50% of the data falls into; the mid-value of the data shown by a line, and extreme values out of the box for the percentage loss are shown with lines and dots. In other words, the data is divided into four quarters where the whiskers and dots out of the box indicate variability in the first quarter and the fourth quarter of the data.
The end of the whiskers represents the minimum and maximum values acceptable in the distribution (and thus identify potential outliers). An outlier is an observation point that is distant from other observations, and thus falls outside of the overall trend that is presented in the data.
In box plots, an outlier is a value, which is greater than the end of the third quarter by more than 1.5 times the distance between the first and third quarter of the data. Outliers are represented with dots.
The FLI methodology uses a simplified food supply chain:
- Harvest/Slaughter: Refers to the act of separating the food material from the site of immediate growth or production.
- On-farm post-harvest/slaughter: All operations and activities carried out by the farmer on the farm, until the product is consumed directly on the farm or exits the farm gate. This may include activities the farmer does directly on the farm, i.e. at home or at the packinghouse, but the product is still owned by the farmer and is under the farmer’s handling.
- Transport: The movement of agricultural commodities, inputs, and machinery from the production site to storage, processing units, or markets, acting as a crucial link in the food supply chain. It includes both traditional (manual/animal) and mechanized methods, aimed at timely, efficient distribution.
- Storage: This takes place when the product is intentionally placed in a specific location to protect it from adverse conditions or to hold it until it can be marketed.
- Wholesale markets: These markets mainly serve traders and other intermediaries. They may be supplied by purchasing or assembly centres in the rural areas or directly from farms, either by traders or by large farmers. Transactions are predominantly handled by traders although many wholesale markets incorporate “farmers markets”, which give farmers the opportunity to sell their goods directly to retailers. Some markets also allow traders to sell to retailers “off the back of the truck”.
- Processing: The set of manufacturing processes that transform raw materials—crops, livestock, or fish—into intermediate or final products.
- Retail markets: These are markets directly serving consumers mainly in urban areas, such as provincial, towns and city centres. Although primarily retail, they may have some semi-wholesale functions, particularly if they allow farmers to trade. In that case, they are often called farmers markets. This operation is very typical in developing countries, but there has also been a strong global trend to create farmers markets to sell specialized produce, such as organically grown fruits and vegetables (FAO, 2014).
- Consumption: This can be at household level or within the food service industry.
For some countries and commodities additional stages or activities have been included as they are critical loss points that were the objective of the study:
- Pre-harvest/Pre-slaughter: Constitutes the time frame between maturity and harvesting.
- Grading: The process of sorting agricultural produce into distinct, uniform lots based on pre-defined quality standards—such as size, shape, color, weight, and maturity—to facilitate marketing and ensure consistent consumer quality.
- Trader: An individual or entity that buys, sells, and manages the risk associated with physical agricultural commodities (such as grain, coffee, or livestock) or their financial derivatives, aiming to profit from price fluctuations.
- Packing (Packaging): The process of wrapping, containing, or enclosing produce and farm products in suitable materials to protect them from damage, contamination, and spoilage during handling, storage, transportation, and marketing. It facilitates easier transport and storage while maintaining quality and preventing loss of value or volume.
- Distribution: The organized systems, processes, and networks — including transportation, storage, processing, and marketing — that move agricultural products (crops, livestock, inputs) from producers (farmers) to final consumers or users. It acts as a vital link connecting farm production with market demand.
- Market: The system, place, or mechanism — physical or virtual — where buyers and sellers exchange farm products, establish prices, and determine supply and demand. It involves the entire value chain from farm-gate to consumer, including assembling, grading, storing, transporting, and selling agricultural products.
- Food services: Consumption at food distribution centers like restaurants, hotels and canteens.
- Households: Consumption and the household level.
- Export: The sale and shipment of agricultural commodities, livestock, and food products from one country to another, primarily designed to generate income, utilize comparative advantages, and meet foreign demand.
- Whole supply chain: An aggregation of the entire food supply chain.
Data has been categorized in main data collection methods.
- The Expert Opinion method consists in the advice, belief or judgment given by an expert on the losses.
- Case Studies examine food losses of specific value chains, products, persons, or groups over a period. Case studies are often limited to a small set of participants, which may or may not be representative of the overall population.
- Controlled Experiment is a method of data collection where loss measurement is done in a fully controlled environment either in the field or laboratory. It includes direct measurement of a pre-sampled area of production and follows the local harvesting and post-harvest handling practices on-farm and off-farm. They include crop cutting experiments, field trials and laboratory trials. This is a cost-intensive measurement that can benchmark interview-based surveys.
- Modelled Estimates refer to estimation of losses via statistical models.
- Survey: survey-based estimates apply where a survey and representative sample were designed to collect loss data.
- Census data refer to loss data collected in the framework of Agricultural Censuses. In a census all the units in a population are surveyed in order to describe the agricultural census including to establish loss levels.
- FAO’s Annual Agricultural Production Questionnaires are the tool used by FAO to collect official data from the countries annually. Loss data are collected with production and other utilizations of the commodities and reported in FAOSTAT in the Food balances domain.
- Literature Review: this applies to data gathered from a peer reviewed article that is itself a review of other articles, reports and journals on FLW. The reference year is the year the article was published, which may differ from the original reference period of the data.
- FLW Protocol: this is a reporting tool developed by the WRI and other stakeholders where different entities can report on loss and waste using a variety of methods. Entities can range from companies and cities to entire countries.
- National Accounts: loss data or estimates from official sources and systems that have been used to compile national accounts but may not have been reported in the Annual Agricultural Production Questionnaires.
- SDG 12.3.1.a - FLI questionnaires: food loss percentages reported by countries to the FAO within the SDG official reporting process.