Business case for category affinity

A visitor’s activity in one session, such as which category they view most often, can be used for targeting in subsequent visits. Each category page a visitor views during a session is captured, and his or her “favorite” category is calculated based on a recency and frequency model. Then, every time the visitor returns to the home page, the hero image area can be targeted to show content related to that user’s favorite category.

Example of using category affinity

Suppose you sell musical instruments online and want to target sales promotions on bass guitars to visitors who have already expressed interest in guitars in the past. Using category affinity, you can create offers that display only to visitors with this category affinity.

Category affinity algorithm

The category affinity algorithm works as follows:

  • Ten points for the first category viewed
  • Five points for each category clicked after the first
  • When a new category is clicked, 1 is subtracted from all previously clicked categories
  • If a category was already clicked (seen), clicking it again does not subtract 1 from all other categories
  • If a sixth new category is clicked, the lowest scored category of the first five categories is dropped from the calculation
  • At end of session, divide all values by 2
NOTE
When several categories are passed within a single mbox call, the order of categories in categoryAffinities is not guaranteed. An arbitrary category is recorded first and receives a score of 10.

Example: category affinity algorithm

For example, viewing the mens-clothing category, then accessories, then jewelry, then accessories again in a session results in affinities of:

  • accessories: 9 (+5 – 1 + 5)

  • mens-clothing: 8 (+10 – 1 – 1)

  • jewelry: 5 (+5)

When the session ends, and the user later returns to the site, the scores are halved:

  • accessories: 4.5 (9/2)

  • mens-clothing: 4 (8/2)

  • jewelry: 2.5 (5/2)

Assuming the user then views, in order, jewelry, accessories, beauty, shoes, and womens-clothing:

  • accessories: 6.5 (4.5 + 5 – 1 – 1 - 1)

  • womens-clothing: 5 (+5)

  • jewelry: 4.5 (2.5 + 5 – 1 – 1 - 1)

  • shoes: 4 (+5 – 1)

  • beauty: 3 (+5 – 1 - 1)

  • mens-clothing is dropped after the final click of womens-clothing as the lowest-scoring category with a score of 1 (4 – 1 – 1 - 1)

When the session ends, and the user later returns to the site, the scores are halved:

  • accessories: 3.3 (6.5/2)

  • womens-clothing: 2.5 (5/2)

  • jewelry: 2.3 (4.5/2)

  • shoes: 2 (4/2)

  • beauty: 1.5 (3/2)

Use category affinity for targeting

The following sections contain information to help you use a category affinity audience for targeting in an activity.

Create an audience to use category affinity

  1. From the Audiences list, click Create Audience.

    Or

    To copy an existing audience, from the Audiences list, hover over the desired audience, then click the Copy icon. You can then edit the audience to create a similar audience.

  2. Type a descriptive audience name.

  3. Click + Add Rule > Visitor Profile.

  4. From the Visitor Profile drop-down list, select Category Affinity.

    Visitor Profile > Category Affinity

  5. Select the desired category:

    Category Affinity > Category

    Categories include:

    • Favorite Category
    • First Category
    • Second Category
    • Third Category
    • Fourth Category
    • Fifth Category

    The “Favorite Category” and “First Category” options are equivalent.

  6. Choose the Evaluator:

    • Contains (case insensitive)
    • Does Not Contain (case insensitive)
    • Equals
  7. Specify each new value in a separate line (for example, “shoes”).

  8. Click Save.

Use the category affinity audience in an activity

You can use category affinity audiences in any activity. During the three-step guided workflow, on the Target step, choose the desired audience.

Previous pageUse profile scripts to test mutually exclusive activities
Next pageCustomer attributes

Target


Adobe Target Maturity Webinar Series

Adobe Customer Success Webinars

Tuesday, Feb 4, 4:00 PM UTC

Adobe Target innovations, including GenAI, and best practices on AI-powered personalization and experimentation at scale.

Register