This document describes the baseline source file format and field requirements for most advertising and selling channels. We pick up your tab- or comma-delimited product source file (.txt or .csv) via FTP or HTTP from your servers or ours. We can also import XML files generated according to either Google's XML specification or the MyDataFeed’s XML schema.

  Download PDF version 

Using a Current Shopping Feed as Your Source File

If you already have an existing feed file, such as Google Shopping product feed, then you can most likely use it as your source file. You will simply need to host the file at an HTTP link or upload it to FTP.

NOTE: Once you import your source file, DO NOT change the file name and format. You may add new fields/columns to your source file (up to 219 fields/columns in total), but you should not rename or remove existing fields unless they are not being used in any of your feeds. Renaming or removing an existing field that is being used in your feeds can cause listings to be removed from your advertising/selling channels.

Sample source files 

• Products.csv
• Products.xml (MyDataFeeds format)
• Products.txt

Comma/Tab delimited file format 

• Source files must be provided in a UTF-8 encoded format
• The first row must contain the field/column names separated by tabs or commas
• Each product/SKU must be on a separate row/line within the file

You must make sure that no line-breaks are contained within the field content, only between products/SKUs. Line-breaks within product content may cause the product line/row to be parsed incorrectly and fail to import.

XML file format 

You can also import XML files generated according to Google's XML or GoDataFeed's XML schema. See below for more information about our own XML schema.

Required/Suggested Fields 

Below are the required and suggested fields/columns for creating your source file. These fields represent the most common product fields found across the most popular shopping sites. You may add more fields to your source file as needed, however our system can only import 219 fields. Please note, our sample files contain more example fields than are listed below.

NOTE: Field names do not have to be named the same as listed below.

MyDataFeeds XML file format 

The MyDataFeeds XML file format is ideal for platform integrations, although it can be used for a single store import. It is also the preferred format for large product set imports for stores with tens of thousands of SKUs, since it has a built-in paging mechanism to make the import process more efficient.

There are three XML node sections to this file’s schema, Fields, Products and Paging; each is discussed below.

Fields node

The Fields node section lets our system know what product attributes to expect for your import. This node contains multiple child Field nodes (up to 219), each representing a product attribute to be imported from your system. Each Field node must contain a name attribute that contains the field name as a value.

Example Fields Node

Products node

The Products node section lets our system know what product attributes to expect for your import. This node contains multiple child Product nodes, each representing a product in your system. Each Product node must contain child nodes named according to the Fields node section, all fields from the Fields node section must be present in each Product node and the content within each product field node must be contained in <![CDATA[ ]]>  to avoid parsing errors caused by HTML tags within the content.

Example Products Node

Paging node (optional section)

The Paging node section allows our system to relay paging calls to your system, providing an efficient mechanism for large product set imports and eliminating server timeouts on large XML files.

The Paging node contains three child nodes

• Start – This is the starting index of the first product to be returned
• Count - The number of products to return
• Total – The total products to be returned for this import

You must populate each of these nodes as calls are made to your system.

Example Paging Node

Our system will make calls to your system via HTTP and will pass query string parameters that indicate which product page to return.

The HTTP call will be similar to the following:

http://www.MyStore.com/MyWebPage?start=1&count=100

To retrieve the second page of a 1000 products file the call will be similar to the following:

http://www.MyStore.com/MyWebPage?start=101&count=100

Suggestions

If you have over 25,000 items, the ideal way to import is by FTP, keeping in mind that:

• CSV (comma delimited) or TXT (tab delimited) is better over XML

• Select our FTP location for improved connectivity

• ZIP the Source File for a faster upload time 4