Mapping and Properties transformation
Last updated
Last updated
In some case, you would need, for a specific destination, to transform your event before it is sent to the destination. Properties transformation support currently renaming or deleting properties.
Enter on the left the property name you want to add or delete. Then enter on the right the property name to map (or leave empty to remove the property)
Deployment in progress for all destinations
The smart mapping feature shows the mapping between standard events/properties and destination partner properties.
Ex: Standard property: value ⇒ Partner property: revenue
If needed, you can modify this mapping, with custom properties for example.
On settings configuration, you can simply enter values with the text input.
Use brackets button {{ }} to enter a dynamic value, like a property name. Without brackets, the value is considered as a string.
Examples:
{{price}}
corresponds to the property price, the output will be "12,30" for example.
price
corresponds to a static value, the output will be "price".
{{price}} excl.taxes
corresponds to the property price concatenated with the string 'excl.taxes', the output will be “12,30 excl.taxes” for example.
If you click on the T icon, you can switch to functions mode.
Use no-code formulas to transform properties. You can use more than 25 formulas (list here).
In settings, functions could be useful to define a complex value, for example a concatenation of 2 properties to create a unique ID required by the destination.
Coming soon
These functions are also available in the smart mapping feature, to easily transform properties before to send to the destination.
For example, if a destination requires MD5 email, you can select the email in clear and apply the MD5 formula to turn the clear email into a MD5 email, only for this destination (if it is for all destinations, please use our Data Cleansing feature).
Other example, if a destination requires a price with "." (ex: 12.30), but in your database prices are stored with a "," (ex: 12,30), you can transform the data with a dedicated formula to turn the "," into a "."