- Optimizely Feature Experimentation
- Optimizely Full Stack (Legacy)
- Optimizely Web Experimentation
- Optimizely Performance Edge
This feature is only available in Accelerate and Scale for MAU pricing models. Contact your customer success manager to discuss options for upgrading.
You can target experiences to specific audiences based visitor behaviors in real time, and create audiences based on what visitors do on your site.
In Optimizely Web Personalization, you can use behavioral targeting to deliver custom experiences to specific groups of visitors on your site based on their behaviors.
For example, suppose you want to behaviorally target an audience called "Luxury Travelers." Luxury travelers are visitors who purchase bags over $200 in the last 30 days.
Behavioral targeting for this audience in Optimizely Web Personalization takes only a few steps:
- Create a page in Optimizely Web Experimentation that targets product details for bags.
- Set up an add-to-cart event for bags.
- Add price and subcategory tags to tell Optimizely Web Experimentation when a visitor purchases bags of a certain value.
See Prepare Optimizely Web Experimentation for your site for more information.
Target this specific behavior by creating a Luxury Buyers audience in Optimizely Web Personalization.
The following example shows how to set up a Luxury Buyers audience for visitors who have made a purchase priced at over $500 in the last 30 days:
- Go to the Audiences dashboard. Click Create New Audience.
- Name the audience Luxury Buyers.
- Under Visitor Behaviors in Audience conditions, select Add to cart click.
- Add filters based on the Price tag.
- Click Save Audience.
Behavioral events and browser local storage
Optimizely Web Experimentation stores events triggered by a visitor in their browser's local storage. This lets behavioral audience conditions be evaluated in real-time as soon as the snippet is downloaded from the CDN and without requiring a separate targeting request.
- Directly access values stored in event tags and use them in your variation code. For example, you might configure an event which tracks product detail page views. You could capture the URL of the product image using a tag. Then, create an Optimizely Web Personalization campaign on the homepage which uses the behavior API to access the product image tag from the most recent product detail page view event and displays the image in a promotional message.
Optimizely Web Experimentation retains up to 1,000 events per visitor in local storage. If a visitor triggers more than 1,000 events without clearing their local storage keys, Optimizely Web Experimentation drops old events to make room for new events.
Behavioral audiences and campaign activation timing
Behavioral audiences are evaluated immediately each time the Optimizely Web Experimentation snippet activates, typically on initial page load. This means that only those events that occurred prior to page load are used to calculate a visitor's membership in any behavioral audiences used by experiments and campaigns which subsequently activate on that page.
Suppose you create an audience called Article Junkies which uses the condition Visitor converted on Viewed Article at least 2 times in the last 7 days. You use this audience to target a personalized experience which is displayed on all article pages. You should expect that under these conditions, the campaign will not activate until a visitor views their third article. Why? Here is how it works:
- Visitor views an article page and triggers a Viewed Article event.
- Visitor views a second article page. When the page loads, Optimizely Web Experimentation checks behavior, sees one Viewed Article event, and decides the visitor is ineligible for the personalization campaign. A second Viewed Article event is now triggered.
- Visitor views a third article page. When the page loads, Optimizely Web Experimentation checks behavior, sees two Viewed Article events, and decides that the visitor is eligible for the personalization campaign. A third Viewed Article event is triggered