What is it
Event definitions represent the raw data captured from various interactions and touchpoints between people and a business, for example. These events are the building blocks for creating more structured objects and analyzing customer behavior. By applying transformations, you can convert event data into meaningful objects, enabling a deeper understanding of who your customers are and how they interact with your business.
-
Definition: Actions triggered by people or AI that are recorded in the CXP.
-
Examples: Website visits, product purchases, email openings, ad clicks.
-
Characteristics:
-
Timestamps: Usually associated with a specific date and time.
-
Contextual Information: Includes data like location, device, and specific actions taken.
-
Usage: Used to analyze customer behavior and gain insights into interactions and engagement.
-
Tips & tricks
Avoid making a new event definition for everything you want to add to the CXP, re-use existing event definitions that match the data you want to enter into the CXP.
For example wen uploading files into the CXP you might want to add new people through a list, re-use an existing definition like 'Update Person' for example to do this. This way your CXP is less cluttered with random event definitions you are not using anymore and you avoid creating complex queries where you need to add all kind of different event filters.
How to create an event definition
Events specify the raw data fields that are ingested into the CXP. These fields come directly from various data sources and are used as the foundational elements for building objects.
When creating a new definition select the right type and follow the steps, go to create new definition for more details.
Keep in mind that event definitions:
-
Do not have a unique identifier
-
Does not have an external key
-
Can hold PII data
-
Can hold sensitive data
Use cases
Page View
Description : Captures information when a person views a webpage or screen on a mobile app.
Key Attributes :
-
URL or Page ID
-
Timestamp
-
Visitor ID or Session ID
-
Referring Source (e.g., search engine, social media)
Click Event
Description : Records instances when a person clicks on a specific link, button, or element within a digital interface.
Key Attributes :
-
Element ID or CSS Selector
-
Timestamp
-
Visitor ID
-
Click Location (e.g., coordinates on a screen)
Purchase Event
Description : Logs the details of a person completing a purchase transaction online or in-store.
Key Attributes :
-
Transaction ID
-
Customer ID
-
List of Products or SKUs
-
Purchase Amount
-
Payment Method
Form Submission
Description : Captures data when a person submits a form, such as a sign-up or contact form, on a website.
Key Attributes :
-
Form ID or Name
-
Timestamp
-
Submitted Data Fields (e.g., name, email)
-
Visitor ID
Email Opened (sent, clicked etc.)
Description : Indicates that a person has opened an email sent from a marketing campaign.
Key Attributes :
-
Email ID or Campaign ID
-
Timestamp
-
Recipient ID
-
Open Device or Client
Cart Abandonment
Description : Occurs when a person adds items to a shopping cart but does not complete the purchase.
Key Attributes :
-
Cart ID
-
Customer ID
-
List of Abandoned Products
-
Timestamp of Last Cart Activity
Not transforming events to objects
In some scenarios, businesses may choose to send only events to the CXP without creating persistent objects (e.g., customer profiles, companies, or households). This approach is suitable for use cases where the focus is on tracking interactions and one time interactions without the need to build a unified customer profile.
Use Case: Anonymous Behavioral Tracking for Website Analytics
Scenario
A retail company wants to track anonymous user behavior on its e-commerce website to analyze trends, optimize the user experience, and improve marketing campaigns. However, the company does not want to create persistent customer profiles (objects) in the CXP because:
-
Many users are anonymous and do not log in or provide identifiable information.
-
The company is focused on aggregate insights rather than individual-level personalization.
-
Data privacy regulations (e.g., GDPR, CCPA) require explicit consent to store identifiable customer data, which the company does not yet have for all users.
What Is Sent to the CXP
The company sends events to the CXP, such as:
-
Page Views: Tracking which pages users visit.
-
Product Views: Tracking which products users view.
-
Add to Cart: Tracking when users add items to their shopping cart.
-
Purchases: Tracking completed transactions (if identifiable information is not available).
-
Search Queries: Tracking what users search for on the website.
Each event includes contextual metadata but does not include unique identifiers like email addresses or customer IDs.
Limitations of Sending Only Events
No Unified Customer View
Without creating objects, the company cannot build a 360-degree view of individual customers or track their interactions across channels.
Limited Personalization
The company cannot deliver personalized experiences (e.g., tailored recommendations or targeted marketing) without persistent customer profiles.
Session-Based Analysis
Insights are limited to session-level or device-level tracking, which may not provide a complete picture of user behavior over time.
No Historical Context
Events are not linked to a persistent object, so there is no way to track a user's journey or history across multiple sessions.
When to Use This Approach
Sending only events without creating objects is suitable for the following scenarios:
Anonymous Users:
When most users are anonymous, and no identifiable information is available (e.g., email, phone).
Aggregate Insights:
When the focus is on macro-level trends rather than individual-level insights.
Data Privacy Concerns:
When the company wants to minimize the collection of identifiable information to comply with data privacy regulations.
Early-Stage Implementation:
When the company is in the early stages of implementing a CXP and wants to start with event tracking before building customer profiles.