This article serves as a guideline of items to consider when creating audiences using results from Tealium Predict.

In this article:

Choosing Thresholds

After a model is deployed and making predictions, a common next step is to use your predictions to create one or more audiences. The prediction values in an Tealium AudienceStream output attribute for any model are decimal numbers in a defined range (0 - 1). The typical approach is to choose one or more thresholds or cutoff points to define which portion of that prediction range you want included in your audience.

For example, if your model is predicting a likelihood to purchase in the next 3 days, you may want an audience of only the visitors who are most likely to purchase. To achieve this, you can create an audience using a rule that defines the threshold as 0.8 as an example. In this scenario, the audience only includes visitors with values greater than or equal to 0.8.

There is not one objectively correct threshold value for all audiences, as different audiences serve different business goals and since there are usually substantial variations between models for different target behaviors or among different datasets.

Sensitivity and Specificity

To choose the best threshold, consider the potential trade-offs of choosing a threshold that is relatively high or low. For example, a trade-off between ending up with a larger audience that is less accurate or a smaller audience that is more accurate. In statistics terminology, this is known as a trade-off between Sensitivity and Specificity.

  • Sensitivity
    Sensitivity represents the True Positive Rate or how well the model correctly identifies all visitors who will purchase (using purchase as the behavior your model is predicting).
  • Specificity
    Specificity determines how well a model identifies false values. Specificity is the metric that evaluates a the ability of the model to predict true negatives of each available category. The True Negative Rate shows how well the model correctly identifies all visitors who will not purchase (using purchase as a the behavior your model is predicting).

Output Attribute Scope

With every deployed model, you generate a visit-scoped output attribute. Because the output attribute by default does not persist after each visit, you may want to enrich this attribute to be visitor-scoped. With a visitor-scoped output attribute, you can use additional tools such as the Visitor Lookup Tool or Audience Discovery.

Complete the following steps to enrich an output attribute generated by Tealium Predict:

  1. Navigate to Predict > Deployed Model Dashboard.
  2. Click Model Details for the model associated with the output attribute you want to enrich.
  3. In the model Live Performance screen, click the output attribute.
    predict_output_attribute.png
  4. In the Visitor/Visit Attributes screen, note the name of the output attribute you want to enrich.
  5. Click + Add Attribute.
  6. Select the Visitor scope and click Continue.
  7. Select the Number attribute and click Continue.
  8. Configure the attribute with the following settings:
    1. In the Title field, enter in a name for the Visitor-scoped Number attribute. For example, "Most recent score for {your output attribute name}".
    2. In the Type field, select Decimal (default).
    3. Click Add Enrichment and select a Set Number enrichment.
    4. Set the enrichment to the name of your Visit-scoped output attribute.
    5. Set this enrichment to run at the end of each visit by selecting Visit ended in the WHEN field.
      predict_visitor_scoped_attribute_properties.png
  9. Click Finish.
  10. Save and publish your profile.
Public