Cross-Device Analytics transforms Adobe Analytics from a device-centric to a person-centric analysis tool. Using Cross-Device Analytics you can answer questions such as:
- How many people are interacting with my brand? How many and what types of devices do they use? How does their use of multiple devices overlap?
- How often do people begin a task on a mobile device and then later move to a desktop PC to complete the task? Do campaign click-throughs that land on one device lead to conversion somewhere else?
- How does my understanding of campaign effectiveness change if I take into account cross-device journeys? How does my funnel analysis change?
- What are the most common paths users take from one device to another? Where do they drop out? Where do they succeed?
- How does the behavior of users with multiple devices differ from users with a single device?
For more information about Cross-Device Analytics please see the following resources:
Adobe has created an Analysis Workspace template to help new users become familiar with key concepts of cross-device analysis. In this Spark page we will introduce you to the Cross-Device Analysis template and provide a number of tips and tricks.
The Cross-Device Analysis Template
Launching the template
To launch the Cross-Device Analysis template, open Analysis Workspace and click the button labeled Create New Project:
Choose Standard Templates at the top of the dialog box, and scroll down until you see the template labeled, Cross-Device Analysis:
Launching the Cross-Device Analysis template (highlighted here in red)
The Cross-Device Analysis template must be used with a Cross-Device virtual report suite, so Analysis Workspace may prompt you to choose which cross-device virtual report suite you want to use with the template. If you do not currently have a cross-device virtual report suite, information about creating one can be found in the Cross-Device Analytics documentation.
After selecting a virtual report suite, the template will open. The template is divided into multiple panels with a table of contents and introduction at the top. You may click on any item in the table of contents to open a specific panel. Alternatively, you may expand any panel by clicking the < symbol on the right hand side:
Panels within the Cross-Device Analysis template
Each panel within the Cross-Device Analysis template illustrates a particular topic. The sections below provide details on each of these topics.
Panel 1: Special note for members of the Co-op Graph
Journey IQ: Cross-Device Analytics (CDA) integrates with the Adobe Experience Platform Identity Service, utilizing either the Co-op Graph or Private Graph to identify how devices map to people. CDA then leverages intelligence from the graph to create a cross-device view of user behavior. While similar in purpose, the Co-op Graph and Private Graph have a few key differences:
A comparison of Co-op Graph and Private Graph
Because the Co-op graph is available in North America only, you may want to filter your analysis accordingly for report suites that contain data from outside the US and Canada. The Cross-Device Analysis template includes two built-in segments for this purpose:
- Co-op regions. Filters the data to US and Canada only.
- Co-op unsupported regions. Filters the data to all regions except the US and Canada.
Within the Cross-Device Analysis template, the panel labeled Special note for members of the Co-op Graph provides an example of how you may use these two co-op-related built-in segments:
Special note for members of the Co-op Graph
Panel 2: Identification of users
Cross-device stitching in CDA relies on the Co-op Graph or Private Graph to determine which devices belong to each person. To better understand how well the graph is identifying your audience, the Cross-Device Analysis template uses the Identified State dimension in a number of built-in segments and calculated metrics:
- Identified people (dimension and calculated metric)
- Unidentified people (dimension and calculated metric)
- Identified on 1 device (segment)
- People identified on 1 device (calculated metric)
- Identified on 2+ devices (segment)
- People identified on 2+ devices (calculated metric)
- % identified people (calculated metric)
The Cross-Device Analysis template illustrates several ways you can take advantage of these built-in segments and metrics:
Panel 3: Measuring audience size
The panel labeled Measuring audience size uses two metrics and one calculated metric to help you measure your audience size based on the number of people.
- People. The People metric represents the number of people in your audience as calculated in Adobe Analytics based on intelligence provided by the Co-op Graph or Private Graph. Each person represents a group of one or more devices.
- Unique Devices. The Unique Devices metric represents the number of devices being used by the people in your audience.
- Cross-device compression. Cross-device compression is a calculated metric that represents the average amount of stitching that is going on within the cross-device virtual report suite. Compression is calculated as follows:
Compression = 1-(People/Unique Devices)
The higher the cross-device compression, the more cross-device stitching that is occurring. Note that compression can only occur when two conditions exist:
- The graph can identify the person and their device(s)
- The person actually uses more than one device
Using People and Unique Devices you can create useful visualizations such as a Device type overlap Venn diagram, or a Devices-per-person histogram. The Cross-Device Analysis template shows you how to do this:
People, unique devices, and cross-device compression
Compression by day, device type overlap and devices-per-person.
Pro tip: Note that the Venn diagram within the Cross-Device Analysis template uses people-based segments for "Desktop users", "Phone users" and "Tablet users". More information on people-based segments can be found in the section that follows.
Panel 4: People-based segments
The Cross-Device Analysis template includes multiple person-based segments that you can use to analyze your audience in multiple ways. Try applying these segments to any of the panels within the Cross-Device Analysis template to see how they impact your analysis:
- People who use mobile devices only
- People who use desktop devices only
- People who use both mobile and desktop devices
- Phone to desktop conversion
- Mobile device to desktop conversion
- People with 1 device
- People with 2+ devices
Selecting a person-based segment to apply to your analysis
If you inspect these person-based segments you can see how they are built. For example, here is the definition for the Phone to desktop conversion segment:
Let's break this segment down:
- The outermost container is a PERSON container.
- Within this container, the segment specifies two sequential steps: The person uses a mobile phone, then the person uses a desktop device and places an order.
Using a PERSON-based outer container ensures the segment will include cross-device behavior in the segment results. In other words, the segment returns people with interactions that cross a mobile phone and a desktop device. (Note that in Adobe Analytics, Mobile Device Type = Other refers to desktop devices.)
Panel 5: Analyzing the cross-device journey
You may have wondered in the past, "How often do my users move from one device to another?" The Cross-Device Analysis template includes an example cross-device Flow diagram. The flow diagram is configured to use a Person container, ensuring that it includes all cross-device behavior, even if the behavior occurs across multiple visits:
Using the Person container in Flow
The diagram then uses the Mobile Device Type dimension as its basis to illustrate users moving from mobile devices to desktop devices and back again:
The Cross-Device Analysis template also illustrates one way to use Fallout in a cross-device context:
In this example, the Fallout diagram represents a cross-device funnel where the steps are:
- The user interacts on a mobile device
- The user interacts on a desktop device
- The user places an order.
The Fallout diagram is configured to use a Person context, ensuring the funnel will include cross-device behavior even if it occurs across multiple visits:
Using a Person container in Fallout
Pro tip: The segments used in the Fallout diagram within the Cross-Device Analysis template are hit-based segments. With Fallout, you want to specify the exact conditions that must occur for a person to fall through at each step of the funnel.
Example hit-based segment for use in Fallout
Finally, the Cross-Device Analysis template illustrates a cross-device cohort analysis:
Note the criteria for the cohort analysis:
- Inclusion: An interaction occurs on a mobile device
- Return: An interaction occurs on a desktop device
Pro tip: Like Fallout, you should use hit-based segments as the inclusion and return criteria in a cohort analysis so that you are specifying the exact criteria for each step. The cohort will include cross-device behavior because the inclusion criteria is "mobile interaction" and the return criteria is "desktop interaction."
Panel 6: Cross-device attribution
One of the most exciting applications of cross-device analysis comes from combining cross-device analysis with Attribution IQ. The Cross-Device Analysis template includes an example of how you might do this:
Cross-device Attribution IQ
Within the example, we have created three metrics:
- First Touch Revenue. This is the Revenue metric configured through Attribution IQ as first touch.
- Participation Revenue. This is the Revenue metric configured through Attribution IQ as participation.
- Acquisition Rate. This is a calculated metric metric which represents the ratio of first touch to participation:
Acquisition Rate = First Touch Revenue/Participation Revenue
Acquisition Rate represents how well a given item functions as a first touch engagement. In the template, Acquisition Rate is applied against the Page dimension, but you can substitute any other dimension instead.
Configuring a metric column with Attribution IQ
Definition of Acquisition Rate
For more information on using Attribution IQ, see the Attribution IQ documentation.
Panel 7: Other tips and tricks
The final panel in the Cross-Device Analysis template provides you with a few other tips and tricks for optimizing the stitching provided by CDA:
The Cross-Device Analysis template in Analysis Workspace provides new users with an overview of key concepts for performing cross-device analysis. We hope you enjoy the template. Happy analyzing!
Ready for more content?
See Nils Engel and Matt Freestone in their 2020 Adobe Virtual Summit presentation about Cross-Device Analytics: Cross-Device Analytics: Top Tips to Tackle the Complex Customer Journey
Visit adobe.ly/aaresources for a full list of Adobe Analytics Spark pages & other helpful resources.