Data ingestion determines what you can activate, how reliable the identity resolution will be, and how quickly your audiences and journeys can react.
The data you bring in sets the limit on what's possible — You can't target potential customers based on data you don't have.
What gets ingested
Data Activation needs 2 categories of data to build activation flows (audiences and journeys):
Event data
Time-stamped actions describing what someone did
Examples include page view, product view, add to cart, purchase, login, or support ticket opened
Profile data
Attributes describing who someone is in a business context
Examples include customer ID, loyalty tier, subscription status, contract end date, or consent flags
Both categories can come from multiple sources and arrive at different speeds.
Minimum required signals
Even simple use cases require a consistent set of fields.
For events:
Event name: A stable, descriptive name (for example,
product_viewed,purchase_completed).Timestamp: When the event occurred.
Identifier: Customer ID, email hash, or device ID when available.
Key properties: Attributes needed for targeting or journey logic (for example, product category, order value).
For profiles:
Customer identifier: A stable ID that can be matched across events and destinations, such as ad platforms.
Core attributes: The 5-10 attributes that matter most for your activation use cases.
Consent and preference fields: When applicable for compliance and targeting.
The specific fields vary by industry and use case. Start with what you need for your first audiences and journeys for testing, then expand.
Example: web events combined with CRM profiles
The following example combines behavioral data from web tracking with customer context from a CRM:
A visitor browses products on the website.
Web events capture product views and add-to-cart actions.
A CRM sync provides customer status, loyalty tier, and marketing permissions.
Identity resolution ties the browsing behavior to the customer profile when an identifier becomes available.
Audiences and journeys can use both behavior and business context, for example, targeting high-value customers differently from new prospects.
This configuration works because web data explains intent, while CRM data explains relationship and eligibility.
Ingestion methods
Select your ingestion approach based on your use case and available systems. Read more below for example use cases and what to keep in mind when setting up event capture.
Web tracking
Web tracking captures behavioral events from websites and web applications in real time.
Web tracking is most useful when you need to respond to browsing behavior (such as, product views, cart activity, page visits). Use web tracking for detecting intent signals, and also for real-time or near-real-time audience updates.
Note that working with web tracking, there are some limitations on what you can do, including:
Web tracking requires JavaScript implementation on your properties. You may need help from your website team.
Regional privacy regulations and browser restrictions may affect what data can be collected from your customers.
Anonymous visitors can be activated to cookie-based destinations (for example, retargeting). However, any destinations requiring an email address or phone number would need either one as a known identifier.
For details, see Web tracking and event collection.
Server-side events
Server-side events are events sent from backend systems, apps, or services directly using an API.
Server-side events are best used when critical events happen outside the browser (purchases, subscriptions, backend actions). They provide a more reliable delivery than client-side tracking. In some cases, privacy requirements favor server-controlled data collection because it is unlikely to be intercepted by bad actors.
Note that working with server-side events, there are some limitations on what you can do, including:
Due to the complexity of setting up server-side events, they require development resources from your organization to be instrumented.
Server-side events may introduce latency to your data flow, depending on your backend architecture.
You're responsible for ensuring data quality before it's sent. Make sure that the data is useful and valuable to your needs.
For details, see Server-side events and APIs.
Profile ingestion from business systems
Profile ingestion from business systems involves loading customer attributes from CRMs, data warehouses, billing systems, and other sources to the Data Activation platform.
Ingesting profiles from business systems works best when your targeting depends on business context (customer status, contract dates, segments). It's also required for suppression based on relationship data (such as excluding existing customers or churned accounts). In business systems, profile attributes change less frequently than behavioral events.
Note that ingesting profiles from a business system comes with some limitations, such as:
The data is often batch-based, so that data freshness may lag behind real-time events.
For best results, the data requires a stable identifier that matches across systems.
You need to define a clear ownership of data quality at the source.
For details, see Profile ingestion from business systems.