Expand my Community achievements bar.

How to Leverage Event, Lookup, and Profile Datasets in Adobe Customer Journey Analytics





Adobe Customer Journey Analytics (CJA) is built on three types of datasets to gather actionable insights into your customers' journeys. These dataset types, Event, Lookup, and Profile, each play a unique role in capturing and presenting different aspects of customer data.

Event datasets are structured to record detailed interactions at specific points in time, covering every key action that a customer takes on your platforms. In contrast, Lookup datasets serve as reference points, offering metadata context to the recorded event values. Profile datasets, however, concentrate on the individuals behind those actions, combining the data into customer attribute metadata.

Each dataset type serves a unique purpose and provides specific data points critical to understanding customer interactions. This blog post offers an overview of these three CJA dataset types. It highlights their key characteristics and potential applications, helping you understand their functionalities and how they can enhance your CJA instance.


Event Datasets

CJA predominantly utilizes Event datasets, which record user activity data along with corresponding timestamps. These structured datasets form the fundamental backbone of any CJA implementation. They provide a broad, comprehensive view of user activity, thereby offering a holistic understanding of user engagement. Once this baseline data is established, other types of datasets can then be applied to this primary activity data, on a Profile or Record lookup basis.

Event data, which is an essential part of understanding user interactions, encompasses a wide and diverse array of potential point in time interactions. These interactions could range from a variety of different contexts such as website or mobile activity, which can provide key insights into how users are engaging with your platforms or application. Additionally, it can also include online or offline transactional data or point of sale data, which can give valuable information about customer buying habits and preferences. Importantly, it can also cover key behavioral events such as propensity indicators, churn, and attrition, offering deep insights into user behavior and trends. Each row in the Event dataset signifies a unique recorded event, providing a comprehensive overview of all activities. Consequently, this event data serves as a detailed record of all user interactions with your platforms, making it an invaluable resource for enhancing user experience and engagement.

By examining and analyzing events, organizations can obtain valuable insights into user interactions with their platforms, systems, or services. This information can help identify potential issues or challenges, as well as detect emerging trends, patterns, or changes in user behavior. Such insights can highlight opportunities for optimization or activation. This information is invaluable for predicting future user behavior and adjusting strategies accordingly, ultimately informing strategic business decisions and actions.

Event Dataset Key Takeaways

  • Any built-in or custom schema that is based on an XDM class with the "Time Series" behavior can be used. Examples of these include "XDM Experience Event" and "XDM Decision Event." Ad-hoc Schemas are not currently supported for Event datasets in CJA
  • Every row must contain timestamp information which is resolved to the millisecond level.
  • The Person ID declared in the Connections settings must be included in the event row. If it's not present, the record will be skipped during ingestion into CJA.
  • At present, a CJA Connection must include at least one Event dataset.
  • Merging multiple Event datasets can provide a comprehensive view of customer behavior across different contexts.

Lookup Datasets

Lookup datasets are used for data enrichment, enhancing the quality and depth of event data, much like classifications in Adobe Analytics. They provide a retrospective mechanism to supplement initial event data with metadata that wasn't originally included, offering a more comprehensive and detailed view of the events. Data from CJA record lookups is only visible in CJA when mapped to the corresponding matching active record key.

This additional layer of information can prove invaluable when it comes to analyzing complex data patterns, as it allows for a more nuanced understanding of the events. It can shed light on subtle trends and correlations that might otherwise go unnoticed, thereby providing a richer, more detailed picture of the event data.

By offering a more comprehensive understanding, Lookup datasets can be instrumental in generating more thorough insights, which in turn can aid in informed decision-making. This aspect becomes particularly beneficial when dealing with complex data structures or when analyzing data patterns that require a high level of detail and precision.

In essence, the use of Lookup datasets significantly amplifies the value and utility of event data. They serve as an effective tool to not only enrich the data but also to enhance the overall quality of reports and analysis, thereby translating into more accurate and effective business insights.

Lookup Dataset Key Takeaways

  • Lookup datasets use the XDM “Record” class. Ad-hoc Schemas are not currently supported for Record datasets in CJA
  • A specific field needs to be defined for mapping between the Experience Event dataset and the Lookup dataset.
  • Custom Record Lookup Datasets allow for precise lookups on unique keys located in their Event, Profile, or Lookup data. This simplifies data integration and enhances data analysis. Currently, there is a limit of 10 million unique keys per dataset. If this limit is surpassed, only the initial 10 million unique keys will be included in the dataset.
  • Standard (Adobe-Supplied) Lookup Datasets are pre-defined Lookup datasets provided by Adobe, offering standardized and consistent data enrichment options for browser, operating system, and mobile device dimensions.

Profile Datasets

Profile datasets in CJA perform a function that is similar to the role of customer attributes in Adobe Analytics. They serve as a repository for user-level information, which often includes information derived from CRM systems. Data from CJA profiles is only visible in CJA when it is mapped to corresponding person events. These events must also be triggered within a specific reporting timeframe in CJA.

This additional layer of data serves to enrich the event data that is already being collected, providing a more comprehensive and detailed picture of the user. This is particularly valuable as it allows for a deeper understanding of user behavior and preferences, enabling more accurate targeting and personalization efforts.

Another important characteristic of Profile datasets in CJA is their lack of temporal tracking. Instead of tracking state changes over time or being tied to a specific point in time, Profile records in CJA reflect the most recent user data. Operating on a last-value overwrite basis, new data replaces any previous data. This ensures that Profile datasets always represent the most current state of the user's data.

Profile Dataset Key Takeaways

  • CJA Profile datasets are specifically designed for CJA use and are not derived from the "Real-Time Customer Profile"
  • If you establish a Connection incorporating datasets with varying IDs, the reports will display this difference. To consolidate datasets, ensure to use the same Person ID.
  • Profile Datasets allow you to apply a wide range of data to your persons, users, or customers in the Event data, providing a more comprehensive view of your audience.


In conclusion, the three distinct datasets used by CJA - Event, Lookup, and Profile - provide insights into intricate customer behavior patterns. By fully understanding the capabilities and uses of each dataset type, organizations can maximize the benefits of CJA.

Event datasets provide granular data on customer interactions, mapping out every touchpoint in the customer journey, from the initial contact to the final purchase or conversion. This detailed view helps businesses identify key moments in the customer journey and understand the conversion process. Record Lookup datasets complement these interaction data points with extra context, enhancing the understanding of each customer interaction's circumstances. This additional layer of information paints a fuller dimensional picture of the customer journey. Finally, Profile datasets provide a detailed view of individual customers, including their status, attributes, and categories.


1 Comment