Data Collection

Email event data collection

Through our integration with our email service provider partner, Airship captures email event data to measure campaign performance, track engagement, take action on spam complaints, and to handle unsubscribe requests and other email bounce events. This information is used to populate reports, including Performance AnalyticsA customizable marketing intelligence tool that provides access to reports and graphs based on engagement data.. It is also made available for processing outside of Airship via Real-Time Data StreamingA service that delivers user-level events in real time to your backend or third-party systems using the Data Streaming API..

The following is a list of email events that we capture:

Event name	Description
injection	Message is received by or injected into our email service provider partner
delivery	Remote Message Transfer Agent or Mailbox acknowledged receipt of a message
bounce	Remote Message Transfer Agent or Mailbox has permanently rejected a message
spam_complaint	Message was reported as spam by the recipient
out_of_band	Remote Message Transfer Agent or Mailbox initially reported acceptance of a message, but it has since reported that the message was not delivered
delay	Remote Mailbox has temporarily rejected a message
click	Recipient clicked a tracked link in a message
open	Recipient opened a message in a mail client, rendering a tracking pixel at the bottom of the message
initial_open	Recipient opened a message in a mail client, rendering a tracking pixel at the top of the message
list_unsubscribe	Recipient clicked the ‘unsubscribe’ button on their email client
link_unsubscribe	Recipient clicked an unsubscribe link in a received email

To provide the services, Airship uses an email provider Subprocessor. The Subprocessor may also process the following data:

• Email cloud sending data
  • Email address and contact data: name, email address, and other demographic and segment data provided by you as the customer
  • Email sending information: IP addresses, usage data, cookies data, location data, browser data
  • Email content
• Event data as described in this document

The data is used by the Subprocessor to provide the services, including event insights of emails sent to customers, provide support and troubleshooting, and to fulfill its legal obligations.

Recipient details per event

The following is a list of recipient data that we capture for email events that we capture:

Recipient data	Description	Applicable Events
Channel ID	Email channel identifier	All events
Email Address	Email address of the end-user	All events
Event Name	Type of engagement event	All events
User Agent	User agent string returned for email client or web browser	open, initial_open, click
Is Mobile	Indicates if this was a mobile device	open, initial_open, click
Is Prefetched	Indicates if this event was likely prefetched, for example through Apple MPP	open, initial_open, click
Is Proxy	Indicates if the user agent is a proxy server	open, initial_open, click
Agent Family	Agent or client type based on user agent	open, initial_open, click
Device Brand	Device Manufacturer based on user agent	open, initial_open, click
Device Family	Device Model based on user agent	open, initial_open, click
OS Family	Operating system based on user agent	open, initial_open, click
OS Version	Operating system version based on user agent	open, initial_open, click

Tracking for click, open, and initial_open is enabled by default. You can disable open and click tracking per channel in Contact Management. You can disable open and click tracking per message when configuring Sender Information.

Email pixel tracking

Email opens are measured using a transparent pixel image embedded in the message HTML. When an email client loads the image, the request is recorded as an open event. This helps gather data that indicates the email’s effectiveness, aiding in marketing strategy and optimization of future campaigns.

* initial_open is triggered by a pixel positioned near the top of the message.

• open is triggered by a pixel positioned near the bottom of the message.

Email marketers depend on being able to measure campaign success and audience engagement through pixel tracking. Changes in email client behavior, particularly within Gmail and Apple Mail, now influence the accuracy of these metrics.

Prefetching

Tracking pixel prefetching occurs when an email client, or intermediary proxy, requests remote images, including the tracking pixel, before or without a human viewing the message. These automated requests can register as opens and inflate open and unique open counts. In addition, device and location attribution may reflect proxy infrastructure rather than the end device.

Gmail prefetching and Apple MPP

When a user of Google’s Gmail receives an email during their session, whether on web or mobile app, Gmail triggers prefetching in order to download all email images just before display.

The prefetch request originates from a Google IP address and includes a specific user-agent string, which Airship uses to distinguish prefetch opens from genuine opens. An example of this user-agent string is Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/42.0.2311.135 Safari/537.36 Edge/12.246 Mozilla/5.0.

With the release of Apple iOS 15, iPadOS 15, watchOS 8 and macOS 12 Monterey, Apple began offering Mail Privacy Protection (MPP), which hides the recipient’s email address from senders. This prevents the sender from being able to identify the recipient or use pixel tracking. Also, open timestamps from MPP do not represent the actual time a recipient viewed the message. Similar to Gmail, Apple’s proxy servers employ specific user-agent strings.

Most Apple Mail users enable MPP. The inability to discern genuine opens from Apple’s prefetch opens renders open statistics originating from Apple devices unreliable, so monitoring user-agent strings is important for understanding email opens’ origins.

Interpreting prefetched opens

Prefetching makes it hard to know whether an open is from a real user or an automated client. Simply filtering by user agent isn’t reliable since client behavior varies, and excluding prefetched traffic entirely can distort engagement metrics.

Airship addresses this challenge by combining multiple engagement signals rather than relying solely on open events. Our reporting emphasizes stronger engagement indicators, such as link clicks enriched with URL parameters, to provide a clearer view of genuine user behavior.

Airship separates prefetched opens from genuine opens, giving you flexibility in how you analyze your email performance.

In reporting:

• Message reports count prefetched and unknown opens separately from genuine opens. With the Click Map, you can learn how to position key elements where they are most likely to be seen and engage users. See Email Performance in Message Reports.
• Performance AnalyticsA customizable marketing intelligence tool that provides access to reports and graphs based on engagement data. lets you access separate Total and Unique opens using the IsPrefetched dimension. See Email in Performance Analytics to explore this data.

Best practices for analysis:

• Segment or exclude events flagged as Is Prefetched when analyzing performance.
• Prioritize click events and downstream conversion metrics over opens for KPI tracking.
• Use the separated data to understand the impact of prefetching on your specific audience.

Data privacy

Airship makes HTTPS encryption (also referred to as TLS connection) available for data in transit to or from the Service. For more information, see Airship Security Measures . Email data collected by Airship is not transferred to any third parties unless a partner integration is enabled and configured by the application. For information on individual data requests, see Individual Data Privacy Rights.