Unique Events and Unique Dimension Combinations

In this article:

 

About the Unique Events metric

Unique Events are interactions with content by a single user within a single session that can be tracked separately from pageviews or screenviews. Downloads, mobile-ad clicks, video plays, and interactions with gadgets, Flash elements, and AJAX embedded elements are all examples of interactions you might want to track as Unique Events.

Analytics increments the Unique Events metric by 1 the first time during a session that it receives an event with a unique combination of Category/Action/Label, and ignores subsequent events in the session that have the same combination of Category/Action/Label.

The Total Events metric counts each event regardless of the combination of Category/Action/Label.

For example: During the same session, a user clicks the Play button for the same video 5 times:

  • Unique Events = 1
  • Total Events = 5

The following table illustrates how event data is counted across sessions for unique combinations of Category/Action/Label for a single video. The video player is tagged with the Actions Play, Pause, and Stop.

Category/Action/Label Sessions/Events Metric totals
Video/Play/Video A 10 sessions
22 clicks
Unique Events: 10
Total Events: 22
Video/Pause/Video A 8 sessions
16 clicks
Unique Events: 8
Total Events: 16
Video/Stop/Video A 3 sessions
4 clicks
Unique Events: 3
Total Events: 4
    Unique Events: 21
Total Events: 42

About the Unique Dimension Combinations metric

Unique Dimension Combinations counts the number of unique dimension-value combinations for each dimension in a report. This lets you build combined (concatenated) dimensions post-processing, which allows for more flexible reporting without having to update your tracking implementation or use additional custom-dimension slots.

For example, if you have the dimensions Region, Language, and Mobile Device Info in a report, and you include Unique Dimension Combinations as a metric, then Analytics counts the number of times it sees the same combination of dimension values for each row in the report (for example, it counts the number of unique occurrences of United States + en-US + Google Nexus 5X).

Understanding how often particular dimension-value combinations occur versus how often the individual dimension values occur can provide valuable insight into user behavior. For example, if you were marketing an automobile with different paint and performance options that users could configure on your site, understanding which combinations were most popular would let you preconfigure inventory and minimize delivery times for the largest segments of customers.

The following example illustrates the difference between analyzing dimension values in isolation and in combination.

If you look at the individual values for the Paint dimension, then red would be the obvious color to promote.

Paint Sessions
red 100
blue 75
black 75

 

If you look at the individual values for the Performance Pack dimension, then you would promote the comfort option.

Performance Pack Sessions
comfort 100
sport 75
track 75

 

However, if you add the Unique Dimension Combinations metric, you might discover something different: users configured the blue paint and track performance pack together most often.

Paint Performance Pack Unique Dimension Combinations
blue track 75
red comfort 50
red sport 50
black comfort 50
black sport 25

 

With this information, you have the opportunity to set up the car-configuration widget with the blue paint and track performance pack options as the defaults. And you can ship inventory with that configuration in order to get cars in the hands of customers as fast as possible.

Using the metrics in your reporting

Availability in Analytics

Unique Events is available by default in all of the Standard reports.

Unique Dimensions Combinations is available as an additional metric in all of the Standard reports.

Both metrics are available in Custom reports.

When to use each metric

Use Unique Events when you want to focus your analysis on events. This metric deduplicates event data across Category, Action, and Label during a single session, regardless of how you implement events, so that reporting on and analysis of unique events is straightforward.

Use Unique Dimension Combinations when you want to focus on data that is not necessarily associated with events, but is focused on unique results of user behavior. You can use this metric with any combination of dimensions, and analyze different combinations of dimension values.

In general, you probably wouldn’t want to use the metrics together in the same report, but the following example points out how they differ from one another when they’re counting the same data. The example custom reports use the following dimensions metrics:

  • Dimensions: Page, Event Category, Event Action, Event Label
  • Metrics: Unique Events, Unique Dimension Combinations

The dimensions and metrics track the following user behavior:

  • Session 1
    • Views page Product_A.html
      • Fires event
        • Category: Purchasing Activity
        • Action: Add to Cart
        • Label: Size M
    • Views page Product_A.html
      • Fires event
        • Category: Purchasing Activity
        • Action: Add to Cart
        • Label: Size L
  • Session 2
    • Views page Product_B.html
      • Fires event
        • Category: Browsing Activity
        • Action: Add to Wishlist
        • Label: Size L

 

Data when all of the dimensions are included:

Page Event Category Event Action Event Label Unique Events Unique Dimensions Combinations
Product_A Purchasing Activity Add to Cart Size M 1 1
Product_A Purchasing Activity Add to Cart Size L 1 1
Product_B Browsing Activity Add to Wishlist Size L 1 1
Total 3 3

 

Data when one dimension (Event Label) is omitted:

Page Event Category Event Action Unique Events Unique Dimensions Combinations
Product_A Purcahsing Activity Add to Cart 2 1
Product_B Browsing Activity Add to Wishlist 1 1
Total 3 2

 

The Unique Events count is the same in both reports (3) because Unique Events still tracks Event Label, whether or not is is included in the report, and so counts the events that included the Event Labels Size M and Size L as unique events in both reports.

The Unique Dimensions Combinations count changes from the first report (3) to the second report (2) because the Event Label dimension is omitted, and under this circumstance, the metric counts only 2 unique combinations of the dimension values across the 2 sessions.

Was this article helpful?
How can we improve it?