�۶���Ƶ

[Premium]{class="badge positive" title="See what's included in Target Premium."}

Set up A4T reports in Analysis Workspace for Auto-Target activities

The Analytics for Target (A4T) integration for Auto-Target activities uses the �۶���Ƶ Target ensemble machine learning (ML) algorithms to choose the best experience for each visitor based on their profile, behavior, and context, all while using an �۶���Ƶ Analytics goal metric.

Although rich analysis capabilities are available in �۶���Ƶ Analytics Analysis Workspace, a few modifications to the default Analytics for Target panel are required to correctly interpret Auto-Target activities, due to differences between experimentation activities (manual A/B Test and Auto-Allocate) and personalization activities (Auto-Target).

This tutorial walks through the recommended modifications for analyzing Auto-Target activities in Analysis Workspace, which are based on the following key concepts:

Create the A4T for Auto-Target panel in Analysis Workspace

To create an A4T for Auto-Target report, either start with the Analytics for Target panel in Analysis Workspace, as shown below, or begin with a freeform table. Then make the following selections:

Use the Control vs.Targeted dimension to compare the Target ensemble ML model to your control

The default A4T panel is designed for classic (manual) A/B Test or Auto-Allocate activities where the goal is to compare the performance of individual experiences against the control experience. In Auto-Target activities, however, the first order comparison should be between the control strategy and the targeted strategy. In other words, determining the lift of the overall performance of the Auto-Target ensemble ML model over the control strategy.

To perform this comparison, use the Control vs Targeted (Analytics for Target) dimension. Drag and drop to replace the Target Experiences dimension in the default A4T report.

Note this replacement invalidates the default Lift and Confidence calculations on the A4T panel. To avoid confusion, you can remove these metrics from the default panel, leaving the following report:

Figure 2: The recommended baseline report for Auto-Target activities. This report has been configured to compare targeted traffic (served by the ensemble ML model) against your control traffic.

Add experience-level breakdowns of metrics

To gain further insight into how the ensemble ML model is performing, you can examine experience-level breakdowns of the Control vs Targeted dimension. In Analysis Workspace, drag the Target Experiences dimension onto your report, then break down each of the control and targeted dimensions separately.

Figure 3: Breaking down the Targeted dimension by Target Experiences

An example of the resulting report is shown here.

Figure 4: A standard Auto-Target report with experience-level breakdowns. Note that your goal metric might be different, and your control strategy might have a single experience.

Why “Visits” is the correct normalizing metric for Auto-Target activities

When analyzing an Auto-Target activity, always choose Visits as the default normalizing metric. Auto-Target personalization selects an experience for a visitor once per visit (formally, once per Target session), which means that the experience shown to a visitor can change on every single visit. Thus, if you use Unique Visitors as the normalizing metric, the fact that a single user might end up seeing multiple experiences (across different visits) would lead to confusing conversion rates.

A simple example demonstrates this point: consider a scenario in which two visitors enter a campaign that has only two experiences. The first visitor visits twice. They are assigned to Experience A on the first visit, but Experience B on the second visit (due to their profile state changing on that second visit). After the second visit, the visitor converts by placing an order. The conversion is attributed to the most recently shown experience (Experience B). The second visitor also visits twice, and is shown Experience B both times, but never converts.

Let us compare visitor-level and visit-level reports:

Table 1: Example comparing visitor-normalized and visit-normalized reports for a scenario in which decisions are sticky to a visit (and not visitor, as with regular A/B testing). Visitor-normalized metrics are confusing in this scenario.

As shown in the table, there is a clear incongruence of visitor-level numbers. Despite the fact there are two total unique visitors, this is not a sum of individual unique visitors to each experience. Although the visitor-level conversion rate is not necessarily wrong, when one compares individual experiences, visit-level conversion rates arguably make much more sense. Formally, the unit of analysis (“visits”) is the same as the unit of decision stickiness, which means that experience-level breakdowns of metrics can be added and compared.

Filter for actual visits to the activity

The �۶���Ƶ Analytics default counting methodology for visits to a Target activity might include visits in which the user did not interact with the Target activity. This is due to the way Target activity assignments are persisted in the Analytics visitor context. As a result, the number of visits to the Target activity can sometimes be inflated, resulting in a depression of conversion rates.

If you would prefer to report on visits in which the user actually interacted with the Auto-Target activity (either through entry to the activity, a display or visit event, or a conversion), you can:

To create the segment:

Figure 5: Use a segment such as the one shown here to filter the Visits metric in your A4T for Auto-Target report

Once the segment has been created, use it to filter the Visits metric, so the Visits metric includes only visits where the user interacted with the Target activity.

To filter Visits using this segment:

The final panel appears as follows:

Figure 6: Reporting panel with the “Hit with specific Auto-Target Activity” segment applied to the Visits metric. This segment ensures that only visits in which a user actually interacted with the Target activity in question are included in the report.

Ensure the goal metric and attribution are aligned with your optimization criterion

The A4T integration allows the Auto-Target ML model to be trained using the same conversion event data that �۶���Ƶ Analytics uses to generate performance reports. However, there are certain assumptions that must be employed in interpreting this data when training the ML models, which differ from the default assumptions made during the reporting phase in �۶���Ƶ Analytics.

Specifically, the �۶���Ƶ Target ML models use a visit-scoped attribution model. That is, the ML models assume a conversion must happen in the same visit as a display of content for the activity in order for the conversion to be “attributed” to the decision made by the ML model. This is required for Target to guarantee timely training of its models; Target cannot wait for up to 30 days for a conversion (the default attribution window for reports in �۶���Ƶ Analytics) before including it in the training data for its models.

Thus, the difference between the attribution used by the Target models (during training) versus the default attribution used in querying data (during report generation) might lead to discrepancies. It might even appear that the ML models are performing poorly, when in fact the issue lies with attribution.

The exact metric definition, and attribution settings depend on the optimization criterion you specified during activity creation.

Target defined conversions, or Analytics metrics with Maximize Metric Value Per Visit

When the metric is a Target conversion, or an Analytics metrics with Maximize Metric Value Per Visit, the goal metric definition allows for multiple conversion events to occur in the same visit.

To view goal metrics that have the same attribution methodology used by the Target ML models, follow these steps:

These steps ensure that your report attribute the goal metric to the display of the experience, if the goal metric event happened any time (“participation”) in the same visit that an experience was shown.

Analytics metrics with Unique Visit Conversion Rates

Define the visit with positive metric segment

In the scenario where you selected Maximize the Unique Visit Conversion Rate as the optimization criteroin, the correct definition of the conversion rate is the fraction of visits in which the metric value is positive. This can be achieved by creating a segment filtering down to visits with a positive value of the metric, and then filtering the visits metric.

Figure 7: The segment definition filtering to visits with a positive order. Depending on your activity’s optimization metric, you must replace orders with an appropriate metric

Apply this to the visits in activity filtered metric

This segment can now be used to filter to visits with a positive number of orders, and where there was a hit for the Auto-Target activity. The procedure of filtering a metric is similar to before, and after applying the new segment to the already filtered visit metric, the report panel should look like Figure 8

Figure 8: The report panel with the correct unique-visit conversion metric: the number of visits where a hit from the activity was recorded, and where the conversion metric (orders in this example) was non-zero.

Final Step: Create a conversion rate that captures the magic above

With the modifications to the Visit and goal metrics in the preceding sections, the final modification you should make to your default A4T for Auto-Target reporting panel is to create conversion rates that are the correct ratio–that of the corrected goal metric, to an appropriately filtered “Visits” metric.

Do this by creating a Calculated Metric using the following steps:

The complete calculated metric definition is shown here.

Figure 7: The visit-corrected and attribution-corrected model conversion rate metric definition. (Note this metric is dependent on your goal metric and activity. In other words, this metric definition is not re-usable across activities.)

Summary: Final sample Analysis Workspace panel for Auto-Target reports

Combining all of the steps above into a single panel, the figure below shows a complete view of the recommended report for Auto-Target A4T activities. This report is the same as that used by the Target ML models to optimize your goal metric. The report incorporates all the nuances and recommendations discussed in this tutorial. This report is also closest to the counting methodologies used in traditional Target-reporting driven Auto-Target activities.

Click to expand image.

{width="600" modal="regular"}

Figure 10: The final A4T Auto-Target report in �۶���Ƶ Analytics Workspace, which combines all the adjustments to metric definitions described in the previous sections of this tutorial.