Sources - Zeotap docs

What this is

Sources are the entry points that bring customer data into Zeotap CDP. Each source represents one origin of data — a single website, a mobile app, a CRM export feed, a data warehouse table, or a message queue — and streams or batches events and attributes toward a unified customer profile. Zeotap groups sources into nine categories so you can pick the integration that matches how your data leaves its origin system, and every source you create issues its own write key that identifies its traffic on ingestion.

Prerequisites

Before you create a source, confirm:

You have permission to create sources in your Zeotap CDP tenant.
You know the origin system (website URL, app bundle identifier, warehouse project, or file location) and the region where its records live.
Your catalog has the target attributes defined, or you are ready to add them during mapping.

Understand the source categories

Zeotap CDP exposes nine source categories. Pick the one that matches how your data leaves its origin system:

Website Events — capture actions and interactions on a website. Create one source per website. See Website Events.
App Events — track interactions inside mobile applications for a deeper view of user engagement. See App Events.
Flat Files — batch data delivered as CSV or JSON, transferred via SFTP or Zeotap Google Cloud Storage. See Flat Files.
Server-to-Server — server-to-server data transfer through the HTTP API source. See Server-to-Server.
Tag Managers — manage tracking tags and scripts on websites and mobile apps. See Tag Managers.
CRM Data — comprehensive customer records collected through tools such as Salesforce CRM. See CRM Data.
Data Warehouse — centralised structured and unstructured data from warehouses that support analytics. See Data Warehouse.
Customer Engagement Channels — communication touchpoints such as email and push platforms. See Customer Engagement Channels.
Message Queue — payloads or metadata processed in order by another sub-system through a queue. See Message Queue.

Choose one source per origin — a worked example

Create one source for each unique origin. Two websites are two Web JS sources. Three flat files carrying different CRM extracts are three flat-file sources, even if they land in the same folder. A practical scenario: you want to track events on a marketing website.

Choose Website Events as the source category.
Choose Web JS as the source type.
Enter the source name (for example, Website A) and the site URL (for example, a.com).

The result is a source named Website A visible on the Sources listing. When you add a second website later, create a second Web JS source rather than reusing the first — that keeps write keys, catalogs, and delivery traffic separated per origin. Three CRM-derived flat files become three separate sources in the same way:

one file carrying user login information
one file carrying user classifications
one file per website carrying purchase events

Configure a source

Open the Sources module in the Zeotap CDP application and start a new source. Expected result: the source-category picker opens.
Pick the category and source type that match the origin. For a website, pick Website Events and one of Web JS, Web Pixel, or iFrame. For a mobile app, pick App Events and one of Android Native SDK or iOS Native SDK. Expected result: the configuration form for that source type opens.
Enter a short, descriptive source name, then the origin details — website URL, region of upload, and any type-specific fields.
Save the source. Expected result: the source appears in the Sources listing, and its detail view exposes the write key the SDK, snippet, or server-side integration uses to send data to this source.

For flat-file sources, a minimal CSV payload has one row per user, one column per attribute, a header row, country in ISO3, and at least one user identifier:

user_id,email,country,event_timestamp
usr_001,alex@example.com,USA,2024-01-15T10:23:00Z
usr_002,sam@example.com,DEU,2024-01-15T10:24:11Z

Every file uploaded to the source must repeat the same delimiter, header casing, and country encoding it was created with. Mixing formats within one source corrupts ingestion.

Verify the source is receiving data

Once the source is created and the integration is live, open the source’s preview view to confirm events are landing. A healthy source shows recent records in the preview pane, with the expected attribute names and non-null values in the columns you mapped. If no records appear, check:

The origin-side integration is sending to the write key this source issued.
The source is not archived — archived sources pause all future ingestion.
The region of upload matches the region where the records live.

Enable source alert notifications

Zeotap can email designated recipients when a source has an ingestion issue. Configure alerts to notify one or more addresses when any of the following conditions occur:

Alert code	What it means	Where to look
`PULL_FAILURE`	Zeotap could not pull data from the origin system on the last attempt	Source detail → Activity
`INGESTION_FAILURE`	Records reached Zeotap but were rejected during ingestion	Source detail → Activity
`DELAYED_INGESTION_FAILURE`	Ingestion did not complete within the configured freshness threshold	Source detail → Activity
`NO_DATA`	The source received no records in the alert window	Source detail → Activity

Contact your Zeotap customer success representative to enable alert notifications for your org and to set the notification time-of-day, the recipient list, and the failure threshold.

Key features and benefits

Real-time data streaming — streaming sources collect and process data as it arrives, so you can respond to customer actions in the moment.
Customisable data mapping — map origin fields to a standardised catalog schema so downstream audiences and destinations see consistent attributes.
Improved data accuracy — one write key per origin removes cross-source ambiguity when profiles are stitched.
Efficient data management — automated ingestion removes custom-pipeline maintenance work from your team.
Data standardisation — records from every source merge into one catalog held at the user-profile level.

Troubleshooting

A newly added field is missing in preview even though records exist for it — confirm the field was added to the source mapping and that the origin file contains the column with the expected header casing.
Ingestion fails on a flat file that worked before — check that the delimiter, encoding, and header casing match the source’s original definition; a change corrupts ingestion.
The source is archived — unarchive the source to resume ingestion. Zeotap retains the data already ingested before archival. For streaming sources, put the Zeotap pixel or SDK back on your pages so the origin resumes sending.

Next steps

FAQ

Do I need a separate source for each website?

Yes. Create one source per unique origin. Two websites become two Web JS sources so their write keys, catalogs, and delivery traffic stay separated.

What is the write key used for?

Each source issues its own write key. Your integration — the Web JS snippet, a mobile SDK, or a server-side call — includes the write key with every event so Zeotap routes the traffic to the correct source.

What happens to data when I archive a source?

Zeotap retains all data ingested before archival. All future ingestions are paused. For streaming sources, remove the Zeotap pixel or SDK from your pages so the origin stops sending. Unarchiving resumes ingestion.

Can one flat-file source accept multiple file formats?

No. Every file uploaded to a flat-file source must use the same format and delimiter defined at source creation. Mixing formats within one source corrupts ingestion. Create separate sources for different formats.

​What this is

​Prerequisites

​Understand the source categories

​Choose one source per origin — a worked example

​Configure a source

​Verify the source is receiving data

​Enable source alert notifications

​Key features and benefits

​Troubleshooting

​Next steps

​FAQ