Sequential filters

You create sequential filters using the Then logical operator between components, containers and components, or containers. The Then logical operator implies that one filter condition occurs, followed by another.

A sequential filter has some basic functionality and additional options that you can configure to add more complexity to the sequential filter:

• After and within constraints for the Then logic in the sequence filter definition:

• What data to include as part of the overall sequence for the filter definition. Or for a sequence defined as part of a container. By default all matching data are considered. That data is identified by Include Everyone.

  • Select Only Before Sequence to consider only data before the sequence.
  • Select Only After Sequence to consider only data after the sequence.

• What data to exclude as part of the sequential filter definition.

• How to logically group conditions in your sequential filter definition.

Basics

The basics of building a sequential filter are no different than building a regular filter using the Filter builder. You can use the Definition builder to construct your filter definition. In that construction, you use components, containers, operators and logic. A regular filter becomes a sequential filter automatically as soon as you select the Then operator in the main definition or in any of the containers you use within the Definition builder.

Examples

The examples below illustrate how you use sequential filters in various use cases.

Simple sequence

Identify persons who viewed a page and then viewed another page. The event-level data is filtered using this sequence. Irrespective of previous, past, or interim person sessions, or the time or number of page views occurring between the sessions.

Sequence across sessions

Identify persons who viewed a page in one session, then viewed another page in another session. To differentiate between sessions, use containers to build the sequence and define Session level for each container.

Mixed-level sequence

Identify persons who view two pages across an undetermined number of sessions, and then view a third page in a separate session. Again, use containers to build the sequence and define Session level on the container that defines the separate session.

Aggregate sequence

Identify persons who at their first session visited a specific page and then later visited some other pages. To differentiate between the sequence of events, use containers to separate the logic on a Session container level.

Nest a sequence

Identify all sessions where a person visits one page before another page and then have follow-up sessions that involve two other pages. For example, identify all sessions where a person first visits the home page, then a category 1 page and then has other sessions where in each session the category 2 and category 3 page are visited.

After and Within

You can use After and Within the Then operator to define additional time constraints or constraints for Events, Sessions or Dimensions.

Time constraints

To apply time constraints to the Then operator:

1. Select .
2. Select Within or After from the context menu.
3. Specify a time period (Minute, Hour, up until Years).
4. Select the number to open a popup that allows you to type in or specify a number using - or +.

To remove a time constraint, use .

The table below explains in more detail the time constraint operators.

Examples

Some examples of using the time constraints.

After operator

Identify persons that visited one page and then another page only after two weeks. For example, persons that visited the Home page, but the Women | Shoes page only after two weeks.

If a page view for the Home happens on June 1, 2024, at 00:01, then a page view to page Women | Shoes will match as long as that page view occurs after June 15, 2024 00:01.

Within operator

Identify persons that visited one page and then another page within five minutes. For example, persons that visited the Home page and then the Women | Shoes page within 5 minutes.

If a page view for the Home happens on June 1, 2024, at 12:01, then a page view to page Women | Shoes will match as long as that page view occurs before June 15, 2024 12:16.

After but Within operator

Identify persons that visited one page then visited another page after two weeks but within one month. For example, persons that visited the Home page and then after two weeks and within one month the Women | Shoes page.

Any persons hitting the Home page on June 1, 2024 and who are returning to visit the Women | Shoes page after June 15, 2019 00:01, but before July 1, 2019 qualify for the segment.

Event, Session and Dimension constraints

The After and Within constraints allow you not only to specify a time constraint but also an event, session or dimension constraint. Select Event(s), Session(s) or Other dimensions Dimension name. You can use the Search field to search for a dimension.

Example

Below is an example of a sequential filter looking for persons that visited one product category page (Women | Shoes), followed by a checkout page (Checkout | Thank You) within one page.

The following example sequences match or do not match:

Sequence
Page Women | Shoes followed by page Checkout | Thank You
Page Women | Shoes followed by page Women | Tops followed by page Checkout | Thank You

Include

You can specify what data to include in your sequential filter or in a sequential container that is part of your sequential filter.

Everyone

To create a sequential filter that includes everyone, select the option Include Everyone.

The sequential filter identifies data that match the given pattern as a whole. Below is an example of a basic sequence filter looking for persons that visited one product category page (Women | Shoes), followed by a checkout page (Checkout | Thank You). The filter is set to Include Everyone.

The following example sequences match or do not match:

	Sequence
1	Women | Shoes then Checkout | Thank You in the same session
2	Women | Shoes then Men | Shoes then Checkout | Thank You (across different sessions)
3	Checkout | Thank You then Women | Shoes

Only Before Sequence and Only After Sequence

The options Only Before Sequence and Only After Sequence filter the data to a subset before or after the specified sequence.

• Only Before Sequence: Includes all data before a sequence and the first data of the sequence itself. If a sequence appears multiple times as part of the data, Only Before Sequence includes the first hit of the last occurrence of the sequence and all prior hits.
• Only After Sequence: Includes all hits after a sequence and the last data of the sequence itself. If a sequence appears multiple times as part of the data, Only After Sequence includes the last hit of the first occurrence of the sequence and all subsequent hits.

Consider a definition specifying a sequence of a component with criteria identified by B, followed (Then) by a component with criteria identified by D. The three options would identify data as follows:

B Then D	A	B	C	D	E	F
Include Everyone
Only Before Sequence
Only After Sequence

B Then D (occurs multiple times)	A	B	C	D	B	C	D	E
Include Everyone
Only Before Sequence
Only After Sequence

Example

You have defined three version of a sequential filter for site sections. One with the option Include Everyone, one with the option Only Before Sequence, and one with the option Only After Sequence. You named the three filters accordingly.

When reporting on site sections using these three filters, the example output in a freeform table looks like:

Exclude

Filter definitions include all data unless you specifically exclude Person, Session, or Event data using Exclude.

Exclude allows you to dismiss common data and create filters with more focus. Exclude also allows you to create filters excluding specific groups of persons. For example, to define a filter that specifies persons that placed orders and then excluding that group of persons to identify non-purchasers. A best practice is to create rules that use a broad definition rather than trying to use Exclude to target specific persona that match specific include values.

Example of exclude definitions are:

• Exclude pages. Use a filter definition to strip out a specific page (such as Home Page) from a report, create an Event rule where the page equals Home Page, and then exclude the rule. This definition automatically includes all pages except the Home Page.
• Exclude referring domains. Use a definition that includes only referring domains from Google.com and excludes all others.
• Identify non-purchasers. Identify when orders are greater than zero and then exclude the Person.

Exclude can be used to identify a sequence where persons do not be part of specific sessions or perform specific events. Exclude can also be included within a Logic Group (see below).

You can exclude containers, not components.

Examples

See below for examples of using Exclude.

Exclude within

Identify persons who visited one page, did not visited another page, then visited yet another page. You exclude the container using Exclude. An excluded container is identified by a thin red bar on the left.

Exclude at start

Identify persons who visited one page without ever going to another page. For example, people that checked out a purchase without ever visited the home page.

Exclude at end

Identify persons who visited one page but never visited other pages. For example, persons that visited your home page but never any of your checkout pages.

Logic Group

Logic Group enables you to group conditions into a single sequential filter checkpoint. As part of the sequence, the logic defined in the container identified as Logic Group is evaluated after any prior sequential checkpoint and before any following sequential checkpoint.

The conditions within the Logic Group itself may be met in any order. By contrast, non-sequential containers (event, session, person) do not require their conditions to be met within the overall sequence, producing possible unintuitive results if used with a Then operator.

Logic Group was designed to treat several conditions as a group, without any ordering among the grouped conditions. Otherwise stated, the order of the conditions within a Logic Group is irrelevant.

Some best practices to use Logic Group are:

• To group sequential checkpoints.
• To simplify the construction of sequential filters.

Examples

Here are examples on how to use the Logic Group container.

Any order

Identify persons that visited one page, then viewed each page out of another set of pages in any order. For example, persons that visited the Home page, then visited each of the Men page, the Women page, and the Kids page, irrespective of the order.

You can build this filter without a Logic Group, but the construction is going to be complex and laborious. Specify every sequence of pages that the visitor could view. For clarity, only the first container is opened and the other containers are closed . You can derive the contents of the other containers by the titles.

You can use Logic Group to simplify building this filter, as shown below. Ensure you select Logic Group for the container.

First match

Identify persons that visited one page or another page, then visited yet another page. For example, persons that visited the Women page or the Men page, then visited the Checkout | Thank You page.

Exclude And

Identify persons that visited one page then explicitly did not visit a set of other pages, but did visit yet another page. For example, persons that visited the Home Page, did not visit the Men or the Women page, but did visit the Kids page.

Exclude Or

Identify persons that visited one page then explicitly did not visit any page of a set of pages, but did visit yet another page. For example, persons that visited the Home Page, did not visit the Men and the Women page, but did visit the Kids page.

A final example

As a final example, you want to identify persons that learned about a specific product page, without these persons ever touched by your Empower Your Move campaign. And in their first visit to your online store viewed the Home page but did not look further at any fitness (gear) products from the Men category. However, in their next session directly after that, they went to a product page and placed an online order without going through the Home page first.