Prepare source files for import

Before you can import data into Common Data Service, you must create the source data files.

The data source files that you use in an import must be formatted as comma-separated values (.csv), XML Spreadsheet 2003 (.xml), or text files. The use of source files enables the transfer of data from database systems that use different formats into Common Data Service.

A source file may contain data for one entity type or multiple entity types, such as accounts and contacts. For the source files that contain multiple entity data, you must provide a map that includes the <EntitiesPerFile> tag. Set the value of this tag to “Multiple” to indicate that there is more than one entity type in the source file. Add the Dedupe = “Eliminate” attribute to the <EntityMap> tag. This assures that if the file contains duplicate rows for the entity type, a single row is used to minimize lookup related errors.

You can download an example of a data map with multiple entity types from Microsoft Downloads: Look at the MapForSalesForceContactAccount.xml file.

The field values in the source file can be separated by commas, tabs, or other characters that are defined in the ImportFile.FieldDelimiterCode attribute.


Do not use non-printable characters, null (\0) or break (\b), as delimiters for the field values.

The first row in the source file should contain column headings. If you do not include headings, use the ImportFile.IsFirstRowHeader attribute to specify that the first row represents actual data. In this case, default column headings are created with the names Col1, Col2, and so on.

See Also

Import data
Create data maps for import
Add transformation mappings for import
Configure data import
Run data import
Data import entities
Sample: Export and import a data map
Sample: Import data using complex data map