Enriched Events Export

📘

Note

To see if Enriched Events Export is available for your plan, see Optimizely plans

The Enriched Events Export (E3) includes a valuable combination of events attributes:

• Raw metadata (event name, user IDs, etc.) that you pass to Optimizely Experimentation

without additional processing.

• Enriched metadata that Optimizely Experimentation adds, such as experiment ID, variation ID, and session ID.

For more information, see Data Specification.

Use Cases

Optimizely Experimentation's Results/Reports Page helps you understand the performance of your hypothesis with your experiment metrics. Here are several use cases on why you would want to export event data out of the results page:

QA or troubleshoot Optimizely Experimentation’s results

• Get accuracy assurance or troubleshoot event counts.
• Audit the events (including impressions) you sent to Optimizely.

Calculate additional metrics from Optimizely Experimentation's experiment data

• Analyze experiment data via SQL or other data querying methods.
• Create experiment data dashboards using third-party visualization tools.
• Compute advanced metrics such as funnel metrics (this event THEN that event) or compound metrics (this event AND that event) to extend your measurement strategies.

Combine Optimizely Experimentation experiment data with outside data

• Centralize data management by copying source-of-truth experiment data to a third-party data warehouse.
• Join experiment data with other data sources to measure impact on a broader range of external metrics.

Enable the development of applications on top of experiment data

• Create experiment notifications

Service Overview

🚧

Important

We continuously export event data received by your organization to Amazon S3. You can access the event data within one day of receipt (usually available at 13:00 UTC the next day) via Amazon’s API/CLI/SDK using credentials provisioned by Optimizely.

Before exporting the events, Optimizely Experimentation enriches them with valuable information for experiment analysis. For example, the experiment variation an event is attributed to (attribution), and the user session an event occurred in (sessionization).

Click to enlarge