Exclusions

The exclusion configuration allows you to exclude traffic that is not relevant to you or that could distort your analyses when integrated into your data.

The most common cases are:

• Robot exclusion - You detect a robot generating illegitimate traffic and exclude it from the solution

• Internal traffic exclusion - You want to exclude your employees' traffic (by office for example), to prevent your new feature adoption analyses from being skewed

• Test exclusion - You are testing your tagging plan in your production environment, or you want to use Stream Inspector to see how a particular event is collected, but don't want it to be included in your analyses

The exclusion interface allows you to easily configure traffic exclusions, at the event level, based on your criteria.

Events that match the exclusion criteria will be excluded from traffic, emptied of their properties, and quantifiable via dedicated properties.

Create an exclusion

Exclusions are available from Data Management, in the "Configuration" menu.

To create an exclusion, you can click on the "+" icon at the bottom right of your screen. The exclusion creation panel will open.

You can:

• Give a name to your exclusion

• Give a description to your exclusion

Then fill in one or more exclusion criteria:

• Tag parameter - Property name as entered in the tagging (even if not declared as a property)

• User Agent - User Agent as available in the browser

• Event URL - Event URL

• Source URL - Event referrer

• IPs - Address or IP range (not yet available)

Each event that matches the specified criteria will be excluded from traffic.

Before creating an exclusion, make sure to check the scope to which you want it to apply. To all organization sites? To a specific group of sites?

This can be modified using the scope selector in the top right corner of the screen:

Edit an Exclusion

You can edit your exclusions by clicking on the exclusion you want from the exclusion list. You can modify the exclusion name, description, and criteria.

The order of your exclusions is also modifiable and can affect how your traffic is excluded. If two exclusions share a common criterion, the exclusion placed higher in the list will be the one to which the exclusion is attributed.

The order given to exclusions may seem to have little importance. Indeed, an exclusion remains an exclusion.

However, this has an impact on your analysis of excluded events, as the exclusion detail (exclusion_detail) is drawn from the name given to it. See the dedicated paragraph below.

You can also delete your exclusions.

Excluded Events

When an event matches an exclusion criterion, it is automatically excluded.

The exclusion process is as follows:

• The event is renamed to "exclusion.eventname"

• All properties (except "event name") are cleared

• We add three properties:

  • exclusion_type - Will take the value "Custom", as this is a custom exclusion

  • exclusion_cause - Will take the value "Data manager", corresponding to the name of the process used for exclusions

  • exclusion_detail - Will take the exclusion name, as configured in Data Management

Excluded events are not counted in standard metrics. Therefore, they do not impact the number of events, page views, ...

To analyze them, you can create a dataset combining one or more of the above properties with the "events (all)" metric, which takes into account both excluded and non-excluded events.

Test exclusion rules

You can test your exclusion rules from the interface by clicking the "test" button in the top right corner of the screen:

All exclusions will be tested, and if the first one in the list matches your test criteria, it will be mentioned as excluding the traffic.