Create an Automated Personalization activity

From the Target Activities list, click Create Activity > Automated Personalization.

To use the Visual Experience Composer (VEC), click Visual.

To use the Form-Based Experience Composer, select Form. See Form-Based Experience Composer for more information.

(Conditional) Choose a workspace.

Verify or enter the activity URL, then click Create.

The page with the specified URL opens in the VEC.

Click the Name field and type your activity name.

The activity name cannot begin with any of the following characters:

The activity name cannot contain any of the following character sequences:

Modify page elements as explained in Visual Experience Composer options.

You can select multiple images at once from the asset manager. This lets you quickly view the page with each of the images configured for the activity. You can also easily edit text elements in your offers. When you edit an element, bars appear on that element to indicate you have changed it.

Click Manage Content to configure the available combinations.

A dialogue box displays with three options at the top of the screen: Experiences, Offers, and Exclusion Groups.

The Experiences list shows each piece of content selected for the activity and the location that it is assigned to.

You can exclude specific experiences by hovering over the desired experience and then clicking the Exclude icon.

You can batch exclude or include experiences by selecting the checkbox for the relevant experiences and then by clicking the Exclude icon in the top-right corner of the dialog box.

You can filter this list view to see only excluded or only included activities by clicking the Status drop-down list.

(Conditional) Click Offers to select pieces of content and assign them to reporting groups or only allow certain visitors to see certain offers with targeting.

For more information about reporting groups, see Offer reporting groups in Automated Personalization.

(Conditional) Click Exclusion Groups to choose any combination of elements that you want to exclude from the activity.

Although you can create up to 30,000 experiences in an AP test, the algorithm performs its best when fewer than 10,000 distinct experiences are used. This same limit is applied even when the activity has enabled the Disalow Duplicates option.

If you do not currently have any exclusion groups included in your activity, click Create Exclusion Group. You can filter to create a list that shows only the combinations you want to exclude. Name your exclusion group, then click Save.

To edit an existing exclusion group, hover over the group you want to edit, then click the pencil icon.

Click Done when you have finished setting up the content of your activity.

The Targeting step looks familiar if you have used other Target activity types. Here you can select an audience and specify the percentage of visitors who see the control experience by clicking the Custom Allocation drop-down list, then click Next.

The Custom Allocation drop-down list lets you choose from the following options:

• Evaluate Personalization Algorithm (50/50): If your goal is to test the algorithm, use a 50/50 percent split of visitors between the control and the targeted algorithm. This split gives the most accurate estimate of the lift. Suggested for use with “random experiences” as your control.
• Maximizing Personalization Traffic (90/10): If your goal is to create an “always on” activity, put 10% of the visitors into the control. This option ensures that there is enough data for the algorithms to continue learning over time. The tradeoff here is that in exchange for personalizing a larger proportion of your traffic, you have less precision in what the exact lift is. No matter your goal, this option is the recommended traffic split when using a specific experience as the control.
• Custom Allocation: Manually split the percentage as desired.

(Conditional) From the Control drop-down list, select a specific experience to be used as control or select Random Experience.

The control experience provides a comparison to determine how much lift is provided by the automated test.

Automated Personalization always measures performance against a control group. Best practice is to place at least 10% of entrants into the control group. If your goal is to test if the personalization algorithm on the data it is given performs better than no personalization (for example, the randomly served control), a 50/50 percent traffic split between the control and the personalization algorithm is the fastest and most accurate way to achieve this goal. If you want to maximize the amount of traffic that is personalized and you are less concerned with understanding the exact lift your activity is generating, a 10/90 percent traffic split between the control and the personalization algorithm is the fastest and most accurate way to achieve this goal.

Click Next to display the Goals & Settings page.

Configure the activity with the following settings, then click Save & Close.

When you name or rename a metric, the following characters are not allowed: