CSV, JSON, and XML

Choose a format based on what the recipient accepts. First find out which fields, types, and character encoding its system requires. You can deliver the same catalog in several formats through separate exports.

CSV for Excel and accounting systems

Configure the column delimiter, line endings, and BOM for Excel compatibility. If the whole file opens in one column, check the delimiter selected during import. Import product codes with leading zeroes as text so Excel does not change them. The column heading is set in a field's advanced settings.

Nested structures and arrays do not always fit well into a flat table. Agree with the recipient whether it needs one row per product or separate rows for stock at each warehouse. Consider JSON for complex structures.

JSON

JSON preserves nested objects, arrays, numbers, strings, and null values. You can enable indentation for readability and set an optional root container. For example, the path product.sku creates sku inside the product object.

{
  "product": {
    "sku": "A-001",
    "name": "Sample product",
    "price": 1290.5
  }
}

Type matters: 1290.5 is a number, while "1290.5" is a string. Check what the recipient expects. Grouping in format settings and arbitrary code transformations serve different purposes; use dataset processing for complex totals.

XML

Set the root tag and the record tag. Tag names must have a valid form; the server checks them when saving. For example:

<items>
  <item><sku>A-001</sku><price>1290.5</price></item>
</items>

Check how missing values are handled and whether the result matches the receiving system's schema. Changing the root container or tags can break an existing import.

Format options are under Getting data → Format options.