Criteria in Adobe Target are rules that determine which products or content to recommend based on a predetermined set of visitor behaviors. Criteria can be based on popular trends, a visitor’s current and past behaviors, or similar products and content. You can test multiple recommendation types against each other by adding multiple criteria.
The following sections explain more about criteria keys and the recommendation logic you can use for each key. Click the links for more detailed information.
While creating a criteria, you select an industry vertical based on the goals of your recommendations activity.
|Retail/Ecommerce||Conversion resulting in purchase|
|Lead Generation/B2B/Financial Services||Conversion with no purchase|
Other criteria options change based on the industry vertical you select. You can set your default industry vertical on the Recommendations > Settings page or you can specify the industry vertical for each criteria.
The recommendation key you select determines the criteria type. There are several criteria types, which are represented as criteria cards when you set up a Recommendations activity.
The following table explains the various criteria types and their accompanying keys. Click the links for more detailed information about each key.
|Current Page Activity||Recommend items based on what users do on the current page. For example, visitors who view a particular article might want to see other articles from the same category.|
|Custom||Recommend items based on custom attributes.|
|Past Behavior||Recommend items based on how visitors have responded to an item in the past. For example, people who have purchased a particular brand have been more likely to purchase another item from that brand.|
|Popularity||Recommend the most popular items, such as the most popular videos in a related category or the products that have been viewed most often on your site.|
|Recently Viewed Items||Recommend items a visitor has viewed most recently, such as the items a visitor looked at the last time he or she visited your site, or the articles that are trending most highly right now.|
You can also base recommendations on the value of a custom profile attribute.
For example, suppose that you want to display recommended movies based on the movie that a user most recently added to the queue.
Select your custom profile attribute from the Recommendation Key drop-down list (for example, Last Show Added to Watchlist).
Select your Recommendation Logic (for example, People Who Viewed This, Viewed That).
If your custom profile attribute does not directly match to a single entity ID, it is necessary to explain to Recommendations how you want the match to an entity to occur.
For example, suppose that you want to display the top-selling items from a user’s favorite brand.
Select your custom profile attribute from the Recommendation Key drop-down list (for example, Favorite Brand).
Select the Recommendation Logic you want to use with this key (for example, Top Sellers).
The Group By Unique Value Of option displays.
Select the entity attribute that matches to the key you’ve chosen. In this case Favorite Brand matches to
Recommendations now produces a “Top Sellers” list for each brand and shows the user the appropriate “Top Sellers” list based on the value stored in the Favorite Brand profile attribute.
Target Recommendations uses sophisticated algorithms to determine when a visitor’s actions qualify for the criteria set in your activity. The recommendation key determines the recommendations logic options that are available.
|Items/Media with Similar Attributes||Recommends items or media similar to items or media based on current page activity or past visitor behavior.|
|People Who Viewed This, Viewed That||Recommends items that are most often viewed in the same session that the specified item is viewed.|
|People Who Viewed This, Bought That||Recommends items that are most often purchased in the same session that the specified item is viewed.|
|People Who Bought This, Bought That||Recommends items that are most often purchased by customers at the same time as the specified item.|
|Site Affinity||Recommends items based on the certainty of a relationship between items.|
|Top Sellers||The items that are included in the most completed orders. Multiple units of the same item in a single order are counted as one order.|
|Most Viewed||The items or media that are viewed most often.|
|User-Based Recommendations||Recommends items based off of each visitor’s browsing, viewing, and purchasing history. These items are generally referred to as “Recommended for You.”|
If you are running a recommendation and change its criteria, you will lose your reporting data.
You can also use additional known information about a visitor to enhance your recommendations.
All one-day criteria run twice daily. All one-week and longer criteria run once daily. Site Affinity criteria run once daily. Backup criteria run twice daily.
You can view criteria details on a pop-up card by hovering over a card and by clicking the Information icon on a criteria card without opening the criteria.
Click the Algorithm Info tab to view general information about the selected criteria, including its Name, Descriptions, Industry Vertical, Page Type(s), Recommendation Key, Recommendation Logic, and Algorithm ID.
Click the Algorithm Usage tab to view a list of activities that reference the selected criteria. The card lists active, inactive, and draft activities. Click the Live Activities/Inactive Activities/Draft Activities drop-down lists to view the entire list of activities that reference that criteria. You can click the activity link to open the activity for editing.
The Algorithm Usage feature is currently supported for Recommendations activities only. This feature is not currently supported for A/B Test, Auto-Allocate, Auto-Target, and Experience Targeting (XT) activities that include recommendations as an offer.