View reports - A4T FAQ

The Target variable sent to Analytics has a default 90-day expiration period. (Note: this expiration period can be adjusted by Customer Care if needed). As visitors navigate the site throughout this expiration window, they are part of many Target activities, all of which collect in the dimension.

When you segment for an activity to be present in a hit, you get all the experiences that are part of that activity plus any other experiences that are persisting on that hit.

For activities using Analytics as the reporting source (A4T), the goal metric uses the “Increment Count & Keep User in Activity” and “On Every Impression” settings. These settings are not configurable.

For more information, see “While configuring my goal metrics, why can’t I access the Advanced Settings options?” in Metric definitions - A4T FAQ.

There are several options for normalizing metrics in A4T reporting. This metric, also referred to as the counting methodology, becomes the denominator of the lift calculation. It also affects how the data is aggregated before the confidence calculation is applied.

• Unique visitors increment once when a user first qualifies for an activity.
• Visits increment for each session once a user (Unique Visitor) enters an activity, even if the activity is not viewed in subsequent visits.
• Activity impressions increment each time activity content is served. (Measured by Target).

When a visitor views a page that contains an activity, a variable is set for that visitor that contains the name of that activity. See the detailed scenarios below for how each counting methodology compares.

Consider the following:

• The above metrics trigger when a user qualifies for an activity and content is returned from Target. This does not necessarily mean that the user saw the offer. If an activity experience is below the fold and the user does not scroll down the page, then the offer was served by Target but not seen by the user.
• Activity Impressions (measured by Target) and Instances (measured by Analytics) are equal, unless there are multiple mbox calls on the same page in the same activity. This causes multiple Activity Impressions to be counted, but only a single Instance.

For more information, see How to set up A4T reports in Analysis Workspace for Auto-Target activities in Adobe Target Tutorials.

Reports & Analytics applies a same-touch attribution model to “activity impressions” and “activity conversions,” whereas Analysis Workspace displays the raw metrics, which can appear inflated due to persistence of the Target dimension.

To evaluate accurate Activity Impressions and Activity Conversions metrics in Analysis Workspace, ensure that both metrics have Same Touch attribution models applied. Models can be applied by clicking the column settings gear, enabling Non-default attribution models, then selecting Same Touch. Learn more about attribution in Attributes IQ overview in the Analytics Tools Guide.

In other reports, “unspecified” means data did not meet a classification rule, but in A4T this should never happen. If you see “unspecified,” then the classification service hasn’t run yet. It generally takes between 24 to 72 hours for activity data to appear in the reports. Even though the activities do not appear in this report until that time, all visitor data tied to those activities is captured and appear when the classification is complete.

After the classification period, data appears in these reports approximately an hour after it is collected from the website. All metrics, segments, and values in the reports come from the report suite you selected when you set up the activity.

In case classification was done for that activity, and you still see an “Unspecified” row in the report, make sure that the report isn’t using a non-Target metric to display the data. Unless the report is using a Target-specific metric, that “Unspecified” row contains events for calls that are not associated with Target. That row will not contain any Target-associated information (for example, visitors/visits/impressions).

The Target variable sent to Analytics has a default 90-day expiration period. This expiration period can be adjusted by Customer Care if needed. This setting is global for all activities; however, so it should not be adjusted for one case.

You might see Target variables sent to Analytics after the expiration period because the expiration is 90 days, but only if that user never sees another A4T-enabled Target activity. If a user comes back to the site on day 45 and sees another activity, the entire A4T eVar value has its counter reset to 90 days. That means the first campaign from day 1 now could be persisting for up to 45 + 90 = 135 days. If the user keeps coming back, you might get to the point where you see metrics sent to Analytics in your reporting from much older activities. As users delete cookies and don’t return to the site, the numbers in that activity drop, but you can still see them.

This means that activities continue to get page views, visits, and so forth, for up to 90 days after the activity ends for visitors that became part of the activity while it was active. However, if you look at the Activity Impressions metric, you should not see any impressions after the activity ended.

This is normal and expected behavior. The A4T variable functions like any other eVar—the value is associated with the user until it hits the expiration time period (90 days). As a result, if an activity is active for only two weeks, the value is still associated with the user for at least the next 90 days.

Best practice is view reports for that activity only for the time period the activity was live. The dates should be set correctly by default when you view the activity in Analytics, so unless you have manually extended the date this shouldn’t be an issue from a reporting standpoint.

As an example, let’s assume that the A4T variable expires after 90 days and the test is active from January 1 through January 15.

On January 1, the user comes to the site and sees activity XYZ once and has five page views after that. In the next two weeks, the user never returns to the site. The data would look like this for this user:

The user returns on February 1, views five more pages, and doesn’t encounter any more Target activities and the original activity is no longer active. Even though the activity is no longer active, it is still following the user via eVar persistence. The data now looks like this:

The user comes back on March 1 and sees a new activity, ABC. The user also views five pages. Because activity XYZ is still following the user through persistence, and this user then has ABC set, you will see two line items in reporting:

The user then comes back on April 1, views another five pages, and makes a purchase. The 90-day expiration of that first eVar value is reset on April 1, so you see that in reporting. And all the Target activities the user sees receives the credit for the conversion, but the total number of conversions is deduplicated:

Because both experiences were seen before the conversion, they both get “credit” for the order. But, only one order took place in the system and the total reflects that. For Target reporting, because you aren’t putting a Target activity against another activity to see which is more successful, it doesn’t matter that all activities the user saw got credit. You’re comparing the results of two items within the single activity. It’s not possible for a user to see different experiences in the same activity so you don’t have to worry about cross-contamination of order credit.

For more information, see Conversion Variables (eVar) in the Analytics Admin Guide.