Data Quality: Bloomreach Engagement Ghost Sessions

Data Quality: Bloomreach Engagement Ghost Sessions

Data Quality: Bloomreach Engagement Ghost Sessions

What is a ghost session?

You may observe that some users generate brief, activity-free sessions sporadically. This often occurs when a user leaves your website open in a dormant browser tab and their computer enters sleep mode. Periodically, the device may "wake-up" and the dormant browser tab is "pinged" - a behaviour common especially in MacBooks. Consequently, the JavaScript SDK running on the idle tab tracks a session_start and a session_end. This action subsequently records a brief, empty session in Bloomreach Engagement.

(Screenshot) example

This represents a Bloomreach Engagement profile, marked by a series of consecutive Ghost Sessions. It's important to note that this activity is purely fictional.

image

How to identify profiles with Ghost Sessions in your Bloomreach Engagement project

The procedure to identify Ghost Sessions in your Bloomreach Engagement project is straightforward.

Simply generate a report using the specified parameters. This will display customers with an unusually high number of sessions within the past 24 hours, which are, in all likelihood, ghost sessions.

image

How do I quantify the magnitude of this issue in my project?

Predominantly, ghost sessions are tracked within a 24-hour window from the preceding session, and they share the same "path" attribute as the session before them.

With this understanding, you can pinpoint the majority of ghost sessions in your project by following these three steps:

  1. Create a running aggregate that will return the “last session_start path.”
    2. image
  2. Create a running aggregate that will return the “last session_start timestamp.”
    3. image
  3. Create a “session_start timestamp - last session_start timestamp” expression:
    4. image
  4. Create a Ghost Session event segmentation consisting of two groups.
    1. Ghost Sessions with two conditions visible on the screenshot below. This rule will mark any session_start where the URL path is the same and occured within 24 hours of last session_start, in the event segmentation.
      2. image
    2. Regular Sessions (no conditions applied)
      3. image

Consequently, you should anticipate an event segmentation that will display a session division similar to the one illustrated below:

image

Based on our observations, approximately 25-30% of the sessions in your projects are categorized as Ghost sessions.

Bloomreach has devised a solution aimed at curbing the tracking of certain ghost sessions.

Although this method will not entirely eliminate ghost sessions from your project, it's projected to reduce their prevalence by approximately 50%, based on our past experiences.

💡
In the illustrative example presented in this guide, we identified 1.6M session_start events, which implies a corresponding 1.6M session_end events as well. Cumulatively, these account for 3.2M ghost session events. By preventing 50% of these from being tracked initially, you could effectively reduce your event count by 1.6M. These savings would then be excluded from the calculation of your Monthly Processed Events (MPEs).

Fortunately, thanks to the event segmentation we've discussed earlier, you can effectively filter out Ghost sessions from your reports and other Bloomreach Engagement modules.

If you found this post valuable…

Should you find this post insightful, we encourage you to share it with friends and colleagues who might also derive value from it.

Additionally, we offer a complimentary 45-minute consultation if you're interested in delving deeper into the data quality best practices around Bloomreach Engagement.

Thank you for investing your time in this read. We hope the information provided proves beneficial to you!