Configure data sources

Learn how to configure Experience Platform data sources, configure an external data source, and find and use data in a journey.

Transcript
In this video, we will show you how to configure data sources for Journey Orchestration. You will learn how to configure the Experience Platform Data Source to access real-time customer profile data on the platform, how to configure a connection to an external data source in order to access data from other sources, aand how to find and use data in a journey. As an application service built on the Adobe Experience platform, Journey Orchestration can natively read profile attributes and behavioral history from an individual’s real-time customer profile on the platform. There is, however, some initial configuration required to select which profile attributes and behavioral data we want to make available for Journey’s Orchestration in real time. Because the breadth of profile information and the volume of behavioral history that you accumulate for each of your customers can become quite large, it is best for both usability and performance reasons to conservatively indicate which data fields from the real-time customer profile you want to be able to use when creating and executing journeys. You do this by creating field groups, which are groupings of profile attributes and behavioral data that fit together logically and have similar persistence. Out of the box, you’ll find a field group called “Profile Field Group,” which, by default, is based on fields from the union profile schema. As the name implies, this is the union of any XDM profile schemas that you’ve defined on the platform that are enabled for unified profile.
Under Fields, we can select the profile attributes that we want to make available for Journey Orchestration. For example, we might want to include birth dates, gender, name fields, as well as information necessary to communicate with our customers, such as email address, mobile phone number, push token, et cetera. Select the attributes you think you will use, and you can always come back and edit your selections to add additional attributes later if you need them. The cache duration for the profile field group is set to 50 hours by default, which is fine for persistent profile data that rarely changes. For other data that updates more frequently, you can choose a shorter cache duration. For behavioral data, which is stored on the platform as collections of experience events, you can create additional field groups to reference the data that you are interested in. Configuring the field groups is the same. You select a schema, which can be the union of all experience event schemas you have defined on the platform, or select individual schemas and create separate field groups for different types of events. This can make it easier to find and use the behavioral data you’re looking for when creating a journey. Then, you select fields of interest from the chosen schema.
And finally, set the cache duration, which would typically be set much shorter than for more persistent profile attributes, so that you can keep the most up to date behavioral history as the individual progresses through the journey. Then, when you’re done, click Save. In addition to accessing real-time customer profile data on the platform, you can also configure Journey Orchestration to retrieve data from any other data source that can be accessed through a REST API. We’ll demonstrate by setting up a connection to the free OpenWeather API to retrieve current weather data for a given zip code. To set up the connection, we’ll need to know three things: the URL for the API call, including any query parameters used to specify which data to retrieve, an API key for authentication, which you can get by signing up for a free account, and the API response format. We’ll start by clicking Add and giving the new data source a name. Then, paste in the URL, which we find in the OpenWeather API documentation. For authentication, you’ll see options for various authentication patterns, including Basic Authentication, to provide a username and password, an API key, Custom Authentication, which allows you to configure authentication based on JSON Web Tokens, or No Authentication. The OpenWeather API uses an API key. It gets past as a query parameter, so we’ll select API key as the authentication type. The name of the parameter’s APPID, and the value is the API key value that you received when signing up for an account. Finally, we’ll configure the field group to capture the response from the API request. We’ll name it current weather, leave it as GET since this is a GET request, set the cache duration to 10 minutes, which is the frequency with which the current weather data gets updated. For dynamic values, we’ll enter zip, which is the query parameter for passing the zip code for which we want to retrieve weather data. And finally, we’ll paste in the response payload, which we can find in the API documentation. Journey Orchestration will parse the response payload and display the data fields included in the response with the inferred data types. You can double check these data types and adjust if needed. Then click Save, and then save again to save the overall data source configuration. Now that we’ve shown you how to configure data sources for Journey Orchestration, we’ll briefly demonstrate how to use these data sources when building a journey. To do so, I’ll create a new journey, add an event to Start and then a condition where we can demonstrate the use of these data sources in a condition expression. You’ll see in the left rail of the expression editor that you have access to data from the event that initiates the journey as well as field groups of profile data from the platform or any external data sources you configured. For example, I can create a condition that checks whether we have a known name for this individual in their profile.
Or I can use the weather data source to create a condition that checks whether it’s raining. You’ll see the data from this source is grayed out, which indicates that I need to switch the expression editor to advanced mode to work with this data. Like before, we find the field that we’re looking for, which is Main under Weather, and write a condition that it must be equal to rain.
You’ll see an error message saying that you need to provide the zip code parameter in order for Journey Orchestration to look at this data. You do that by clicking on the parameter in the upper right, which opens a new expression field where you can provide the zip code.
You can reference this value from the event context data or from the individual’s profile data, or simply hard code it here.
In summary, being able to access profile and other data in real time is a crucial aspect of orchestrating and delivering real-time personalized customer experiences. You’ve now learned how to configure the experience platform data source to retrieve profile attributes and behavioral data from the real-time customer profile on the platform by creating field groups of the specific profile and experience event data fields that you want Journey Orchestration to be able to access in real time. You’ve also learned how to set up a real-time connection to an external data source by providing the API in point, configuring authentication, and specifying response payload structure. And finally, we’ve given a brief illustration of how to find and use the data from these sources in building a journey. -

For more information, see the About data sources section in the product documentation.

recommendation-more-help
f620e463-d1bd-494c-b200-8abc8052e166