What Is Cross-Device Analytics?
I Am Not My Devices
When users interact with your brand they do so in many ways and on multiple "surfaces" or "devices." They might use a web browser on a PC or mobile device, or they may use a mobile app. In traditional digital analytics, which grew up in data collection based on cookies, each of these surfaces is represented as a unique "visitor." This means each of your human users is represented as a multiple unique visitors.
Here's an example. Suppose Isabelle interacted with your brand in the following way:
Isabelle is three visitors
Using traditional analytics, Isabelle's journey is fractured into three pieces. She is represented as three unique visitors, each of whom used a different device to perform isolated tasks. What is needed is a unified, cross-device view of Isabelle's interactions. Journey IQ: Cross-Device Analytics provides this view.
A Cross-Device View Provides Better Analytics
Having a person-centric, cross-device view of Isabelle's behavior can make a significant difference to your analysis. For example, the traditional visitor-based approach doesn't give you the complete picture of marketing channel effectiveness. Let's look at Isabelle's journey once again, focusing in on which channel receives credit for her product view and for her purchase. We'll use last-touch attribution for simplicity, but the same problem occurs using any attribution model when you divide Isabelle's behavior up into separate visitors. Using the traditional visitor-based view of the world gives very different, even misleading results:
Traditional Analytics vs. Cross-Device Analytics
Notice that with the cross-device view, the email channel receives credit for both the product view and the purchase, which more accurately represents Isabelle's real world experience.
On the remainder of this page, you will learn about:
- How Cross-Device Analytics Works
- Prerequisites For Cross-Device Analytics
- Interpreting Cross-Device Data
- Analyzing Cross-Device Data in Analysis Workspace
How Does Cross-Device Analytics Work?
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. It then leverages this intelligence to create a cross-device view of user behavior. CDA includes unrivaled capabilities and tools to help your enterprise understand multi-device usage and the customer experience across those devices in their interactions with your brand. It sits as a layer below Analysis Workspace to provide deep insight into person-based audience analysis and cross-device attribution, segmentation and journey analysis using powerful tools such as Fallout, Flow, Cohort, Segment IQ and Attribution IQ.
The Cross-Device Virtual Report Suite
CDA is presented through a special kind of cross-device Virtual Report Suite. This allows you to continue to use the original device-based report suite as you introduce cross-device analytics into your organization. Setting up a CDA VRS is easy.
In step one of the VRS builder, choose the report suite that has been configured by Adobe as CDA-enabled:
Choose a CDA-enabled base (source) report suite
Then turn on Report Time Processing and enable cross-device stitching:
Enable report-time processing and cross-device stitching
Finish the VRS setup and save it. The CDA VRS will show up in Analysis Workspace with a special icon next to it as shown below:
Select the CDA VRS in Analysis Workspace
You may create as many CDA virtual report suites as you like on top of the CDA-enabled base report suite.
Sometimes it takes a while for your users to log in and for the Co-op Graph or Private Graph to identify them and map together their devices. CDA utilizes a 30-day lookback window, allowing it to restate a previously unidentified visitor as a person up to 30 days in the past.
How does this help? Recall Isabelle's user journey from the discussion above:
It is possible that Isabelle did not log in until just prior to making the purchase, and that the Co-op Graph or Private Graph did not map together Isabelle's devices until sometime after her purchase. But CDA's 30-day lookback allows CDA to restate Isabelle's past behavior at a person level, providing you with the cross-device view of her journey that you need.
NOTE: Since history can be restated, this means your data may change over time in a CDA-enabled virtual report suite. Keep this in mind as you communicate insights from a CDA-based analysis.
Prerequisites For Cross-Device Analytics
CDA is included with Analytics Ultimate. Beginning in September 2019, Analytics Ultimate customers who meet the prerequisites listed below are eligible to use CDA. The prerequisites for CDA are as follows:
- Your company must be a member of the Adobe Experience Platform Identity Service Co-op Graph, or use an Adobe Experience Platform Identity Service Private Graph.
- You must implement everything required for Co-op Graph or Private Graph including Experience Cloud ID (ECID) and ID syncing with the graph. Note that in addition to technical requirements, the Co-op Graph has other legal and contractual requirements.
- It is not currently possible to use two IMS orgs with a single Private Graph, so you must standardize on a single IMS org. In some cases it is possible for a customer with multiple IMS orgs to use the Co-op Graph in conjunction with CDA.
- The Co-op graph and Private Graph, as well as certain components of CDA are hosted in Microsoft Azure. This means Analytics data is copied back and forth between Adobe's data processing center and Adobe's presence in Microsoft Azure. Some Analytics data will be stored in Azure. Your company must agree to this arrangement.
- CDA requires a "cross-device" report suite. That is, the report suite you use for CDA must include data from multiple different device types or "surfaces" such as desktop web, mobile web, and mobile app. As of September 2019, the server call volume for this report suite must be 100MM server calls/day or less. (Server call volume limits will increase over the coming months.)
- As of September 2019, the Co-op Graph and Private Graph are available in North America only. The schedule for graph presence in EMEA and APAC will be announced at a future time. If you are in those regions we encourage you to start looking at these prerequisites now so that you are ready to go when the graph becomes available.
Interpreting Cross-Device Data
People Not Visitors
Within the CDA Virtual Report Suite, you'll see a few changes. For example, the Unique Visitors metric is replaced by two new metrics: People and Unique Devices. These new metrics give you far better insight into audience size.
People and Unique Devices
In the Segment Builder, the Visitor segment container has been replaced by a Person segment container. Using a CDA VRS, you can create cross-device segments such as:
- People who use more than one device
- People who begin their journey on a mobile device and then later purchase on a desktop device
- Visits where people use more than one device to accomplish a task
Within a CDA VRS, dimensions such as eVars now automatically persist across devices. For example, an eVar that is configured as:
- Allocation: Most Recent (Last)
- Expire after: Purchase
will now automatically persist from one device to another until the Purchase event is fired.
Analyzing Cross-Device Data in Analysis Workspace
Person-Based Audience Analysis
Have you ever wondered how many people are interacting with your brand? Have you wanted to understand how many and what type of devices they use? How does their usage overlap? Using a CDA VRS, you can create cross-device Venn diagrams and devices-per-person histograms.
Person-based audience analysis
With CDA and Analysis Workspace, you can visualize how people are moving from one device to another over time in the Flow visualization. You can see where they drop out in their journey, and where they carry on.
You likely use several Fallout visualizations to analyze how well users make it through a given series of steps prior to achieving success. Did you know your view of those Fallout visualizations is limited when using traditional device-based analytics? In order to successfully "fall-through", the next step must occur in the same browser or app as the previous step. In device-based analytics, you are blind to people who successfully complete the next step on another device.
Not to worry, CDA has you covered. CDA creates the cross-device view that makes Fallout visualizations much, much more useful. After all, what really matters is whether the person ultimately succeeded in their task somewhere.
Cross-Device Attribution IQ
Because CDA creates a layer of cross-device data underneath Analysis Workspace, all your analysis will be flavored with a cross-device perspective. One powerful example is through Attribution IQ. Attribution IQ in Analysis Workspace allows you compare multiple attribution models side-by-side. Using this capability with CDA, you can now compare how different devices contribute to success.
For example, suppose you want to understand how often a mobile phone is the first device used in an interaction that ultimately leads to success. This represents the "acquisition rate" of the mobile phone. CDA + Attribution IQ allows you to do this analysis:
See the Cross-Device Analytics help documentation for more information.
Ready For More Content?
Visit adobe.ly/aaresources for a full list of Adobe Analytics Spark pages & other helpful resources.