This article describes Audience Sizing, how it works, and how to use it to take action on visitors.
Audience Sizing lets you discover and take action on visitors based on their behavior. You specify a date range and rule conditions as input, and your audience data is searched for visitor profiles that match the rules and have had activity within the specified date range. The results are displayed in an interactive Venn diagram showing the number of visitors discovered for each rule condition along with the overlapping conditions, called "segments". A segment can be selected for a one-time use with a connector action. Sending an audience segment to a connector is called a "job".
In order to search visitors, you have to specify a date range and at least one rule.
Step 1: Set a date range
The system will only find visitors that have had activity between the two dates chosen. Visitors that have not had activity within the specified date range or that were purged due to the profile retention time will not be searched.
Step 2: Choose rule conditions
You can select up to three rules. These are the same rules used in visitor enrichments and audience creation.
Important: While the date range specifies the time during which a visitor was active, the rule conditions are evaluated against the current visitor profile, not the profile at the time of activity.
Step 3: Evaluate Results
The search result is presented as an interactive, color-coded Venn diagram consisting of two to three circles, each representing one of the chosen rules. The rules are color coded so that the blue circle represents the blue rule, the red circle for the red rule, and the green circle for the green rule.
Audience Sizing searches are process intensive and take time, especially if you are querying a large number of visitors. As the query is processed behind the scenes, the progress bar will continually reflect the status of the results until the query completes. The progress bar uses sampling of the data set to provide a projection of the final results. This allows you to analyze the results quickly, without having to wait for the entire query to complete.
Visitors are sampled by capping the daily visitor count. The default cap is 100,000. That means if your total visitor population is 1,000,000, the projections are performed on a random sample of a 100,000 visitors. The progress bar displays the progress of this calculation.
The actual query runs on the total visitor population, not just the capped visitors. Likewise, jobs also run on the total number of matched visitors, not just the sampled visitors.
Segments are overlapping regions of the Venn Diagram, representing combinations of the rule conditions. Use your mouse to hover over these segments to display the total number of visitors found. To target a segment with a connector action, check the box next to it and click Next.
You cannot save a segment for later use.
A job is the action taken once you identify your audience segment. You may assign a job to one or more segments at a time. To view the existing job listing, navigate to AudienceStream > Connectors > Jobs.
Things to keep in mind when creating jobs:
Once you have identified a date range and the rules, follow these steps as an example of choosing a segment and creating a job.
Now that the job is created, you can link it to a connector action following these steps:
Now that the job has been linked to a connector you can activate it using these steps:
The job will not start unless the save/publish is successful.
You want to retarget customers who have visited your site in the last week, have not purchased in the last 30 days, have purchased more than $500 in their lifetime, and have a favorite product category of "Jeans".
This example assumes the necessary date selections and rules to identify visitors with a customer's favorite product category, lifetime order value, and days since last purchase.
You want to retarget customers who visited your site during the December holiday season, are currently a VIP Customer, engage most with email, and frequently use coupons.
This example assumes the necessary date selections and rules required to identify visitor’s identified as VIP Customers, engages most with email, and frequently uses coupons.