Merge pull request #5053 from segmentio/predictions-ga

Predictive Traits GA Updates

Author: pwseg

src/_data/sidenav/main.yml

@@ -379,6 +379,8 @@ sections:
         title: Predictive Traits
       - path: '/engage/audiences/predictive-traits/using-predictive-traits'
         title: Using Predictive Traits
+      - path: '/engage/audiences/predictive-traits/suggested-predictive-audiences'
+        title: Suggested Predictive Audiences
 
   - section_title: Journeys
     description: "Learn how to create multi-step Journeys to tailor messages to your users."

src/engage/audiences/predictive-traits/index.md

@@ -3,9 +3,6 @@ title: Predictive Traits
 plan: engage-foundations
 ---
 
-> info ""
-> Predictive Traits is in public beta.
-
 Predictive Traits, Segment's artificial intelligence and machine learning feature, lets you predict the likelihood that users will perform any event tracked in Segment. 
 
 With Predictive Traits, you can identify users with, for example, a high propensity to purchase, refer a friend, or use a promo code. Predictive Traits also lets you predict a user's lifetime value (LTV).
@@ -14,35 +11,34 @@ Segment saves Predictive Traits to user profiles, letting you build Audiences, t
 
 On this page, you'll learn how to build a Predictive Trait.
 
-## Access and build Predictive Traits
+## Access and build a Predictive Trait
 
-To create Predictive Traits, you'll first request demo access, then build a Predictive Trait.
+To create a Predictive Trait, you'll first request access, then build a Predictive Trait.
 
 ![The Predictive Trait builder in the Segment UI](../../images/trait_builder.png)
 
-
 ### Request Predictive Traits access
 
 Follow these steps to access Predictive Trait:
 
 1. Navigate to **Engage > Audiences > Computed Traits**. Select **Create computed trait**.
-2. Select **Request Demo** to access Predictive Traits.
+2. Select **Request Access** to access Predictive Traits.
 
 ### Build a Predictive Trait
 
 Once your Workspace is enabled for Predictive Traits, follow these steps to build a Predictive Trait:
 
 3. In the Trait Builder, select **Predictive Traits**, choose the Trait you want to create, then click **Next**.
-    - Choose **Custom Predictive Goal**, **Likelihood to Purchase**, or **Predicted Lifetime Value**.
+    - Choose **Custom Predictive Goal**, **Likelihood to Purchase**, **Predicted Lifetime Value**, or **Likelihood to Churn**.
 4. (For custom Predictive Goals) Add a condition(s) and event to predict, then select **Calculate**. If you're satisfied with the available data, select **Next**.
 5. (Optional) Connect a Destination, then select **Next**.
 6. Add a name and description for the Trait, then select **Create Trait**.
 
-In the next section, you'll learn more about the three available Predictive Traits.
+In the next section, you'll learn more about the four available Predictive Traits.
 
 ## Choosing a Predictive Trait
 
-Segment offers three Predictive Traits: Custom Predictive Goals, Likelihood to Purchase, and Predicted LTV.
+Segment offers four Predictive Traits: Custom Predictive Goals, Likelihood to Purchase, Predicted LTV, and Likelihood to Churn.
 
 ### Custom Predictive Goals
 
@@ -54,29 +50,39 @@ When you build a Custom Predictive Goal, you'll first need to select a cohort, o
 
 #### Target event
 
-The target event is the Segment event that you want to predict a user's likelihood to perform. Predictions work better when many customers have performed the event.
+The target event is the Segment event that you want to predict. In creating a Prediction, Segment determines the likelihood of the user performing the target event. Predictions work better when many customers have performed the event.
 
 #### Data requirements
 
-Segment doesn't enforce data requirements for predictions. In machine learning, however, data quality and quantity are critical. Segment recommends that you make predictions for at least 50,000 users and choose a target event that at least 5,000 users have performed in the last 30 days. 
+Segment doesn't enforce data requirements for Predictions. In machine learning, however, data quality and quantity are critical. Segment recommends that you make Predictions for at least 50,000 users and choose a target event that at least 5,000 users have performed in the last 30 days. 
 
-You can create predictions outside of these suggestions, but your results may vary.
+You can create Predictions outside of these suggestions, but your results may vary.
 
 ### Likelihood to Purchase
 
-Likelihood to Purchase is identical to Custom Predictive Goals, but Segment prefills the **Order Completed** event, assuming it's tracked in your Segment instance. 
+Likelihood to Purchase is identical to Custom Predictive Goals, but Segment prefills the `Order Completed` event, assuming it's tracked in your Segment instance. 
 
-If you don’t track Order Completed, choose a target event that represents a customer making a purchase.
+If you don’t track `Order Completed`, choose a target event that represents a customer making a purchase.
 
 ### Predicted Lifetime Value
 
-Predicted Lifetime Value predicts a customer's future spend over the next 90 days. To create this prediction, select a purchase event, revenue property, and the currency (which defaults to USD). The following table contains details for each property:
+Predicted Lifetime Value predicts a customer's future spend over the next 90 days. To create this Prediction, select a purchase event, revenue property, and the currency (which defaults to USD). The following table contains details for each property:
+
+| Property        | Description                                                                                                                |
+| --------------- | -------------------------------------------------------------------------------------------------------------------------- |
+| Purchase event  | Choose a target event that represents a customer making a purchase. For most companies, this is usually `Order Completed`. |
+| Purchase amount | Select the purchase event property that represents the total amount. For most companies, this is the `Revenue` property.   |
+| Currency        | Segment defaults all currencies to USD.                                                                                    |
+
+### Likelihood to Churn
+
+Likelihood to Churn proactively identifies customers likely to stop using your product. Segment builds this Prediction by determining whether or not a customer will perform a certain action.
+
+To use Likelihood to Churn, you'll need to specify a customer event, a future time frame for which you want the prediction to occur, and if you want to know whether the customer will or won't perform the event.
+
+For example, suppose you wanted to predict whether or not a customer would view a page on your site over the next three months. You would select `not perform`, `Page Viewed`, and `at least 1 time within 90 days`. 
 
-| Property        | Description                                                                                                                  |
-| --------------- | ---------------------------------------------------------------------------------------------------------------------------- |
-| Purchase event  | Choose a target event that represents a customer making a purchase. For most companies, this is usually **Order Completed**. |
-| Purchase amount | Select the purchase event property that represents the total amount. For most companies, this is the **Revenue** property.   |
-| Currency        | Segment defaults all currencies to USD.                                                                                       |
+Segment would then build the Prediction from this criteria and create specific percentile cohorts. You can then use these cohorts to target customers with retention flows, promo codes, or one-off email and SMS campaigns.
 
 #### Data requirements
 

src/engage/audiences/predictive-traits/suggested-predictive-audiences.md

@@ -0,0 +1,80 @@
+---
+title: Suggested Predictive Audiences
+plan: engage-foundations
+---
+
+Powered by CustomerAI Campaigns, Suggested Predictive Audiences can help you improve customer engagement, drive higher conversion rates, and reduce ad spend.
+
+This page explains what a Suggested Predictive Audience is, how to build a Suggested Predictive Audience, and what each available Audience targets.
+
+## Suggested Predictive Audience basics
+
+A Suggested Predictive Audience is an out-of-the-box Audience template driven by machine learning.
+
+Segment offers [five templates](/docs/engage/audiences/predictive-traits/suggested-predictive-audiences/#suggested-predictive-audience-types) that are prebuilt with [Predictive Traits](/docs/engage/audiences/predictive-traits) like likelihood to purchase and lifetime predicted value. Selecting a template generates a Predictive Audience whose members you can engage in a number of ways:
+
+- [Send an email or SMS campaign](/docs/engage/campaigns/) with a discount code
+- Promote a new product line with a drip campaign
+- Target the Audience members with online ads
+- Send personalized product recommendations
+
+## Build a Suggested Predictive Audience
+
+Follow these steps to build a Suggested Predictive Audience:
+
+1. In your Segment workspace, navigate to **Engage > Audiences**.
+2. From the Audiences tab, select **Go to Predictive Audiences**.
+3. On the Audience you want to build, click **Build Audience > + Add Audience**.
+4. Select the Audience type you want to build, then click **Next**.
+5. On the **Set up requirements** tab, confirm that you have the right events and traits required for the Suggested Predictive Audience, then click **Next**.
+ - If you're missing a required event or trait, Segment prompts you to select it from the dropdown and match it to the required field(s).
+6. Preview your Audience, then click **Next**.
+7. (Optional:) Connect the new Audience to a Destination.
+8. Give your Suggested Predictive Audience a name, then click **Create Audience**.
+
+Your Suggested Predictive Audience is now live.
+
+## Suggested Predictive Audience types
+
+Engage offers five Suggested Predictive Audiences. The following table summarizes the customers each Audience targets and the events and traits Engage uses to build the Audience:
+
+| Audience           | Target                                                                                       | Built with                                                       |
+| ------------------ | -------------------------------------------------------------------------------------------- | ---------------------------------------------------------------- |
+| **Ready to buy**   | Customers who are likely to make a purchase                                                  | `Likelihood to buy` <br> `Order completed`                       |
+| **Long shots**     | Customers who have previously interacted with your brand but aren’t currently engaged        | `Order Completed` <br> `Likelihood to purchase`                  |
+| **High LTV**       | Customers with a high predicted lifetime value                                               | `Predicted LTV`                                                  |
+| **Potential VIPs** | Recently active customers with high predicted lifetime value and high propensity to purchase | `Page Viewed` <br> `Likelihood to Purchase` <br> `Predicted LTV` |
+| **Dormant**        | Inactive customers who are unlikely to purchase                                              | `Page Viewed` <br> `Likelihood to Purchase` <br> `Predicted LTV` |
+
+
+### Audience descriptions
+
+#### Ready to buy
+
+Choose a **Ready to buy** Predictive Audience to target customers who show a high propensity to make a purchase.
+
+Segment builds this Audience with the [Likelihood to Purchase Predictive Trait](/docs/engage/audiences/predictive-traits//#likelihood-to-purchase). Audience members show encouraging engagement and have a likelihood to buy in the top 20th percentile. 
+
+#### Long shots
+
+Choose a **Long shot** Predictive Audience to target customers who have made a purchase but have a middling likelihood to buy.
+
+Segment builds this Audience with the `Order Completed` event and `Likelihood to Purchase` trait. Audience members have completed a purchase but currently have a likelihood to buy somewhere between the 25th and 65th percentile. 
+
+#### High lifetime value
+
+Choose a **High lifetime value** Predictive Audience to target customers that show a high predicted lifetime value.
+
+Segment builds this Audience with the [Predicted LTV Predictive Trait](/docs/engage/audiences/predictive-traits//#predicted-lifetime-value). Audience members are in the top 10th percentile of predicted lifetime value and Segment expects that they'll spend the most over the next 90 days.
+
+#### Potential VIPs
+
+Choose a **Potential VIPs** Predictive Audience to target customers exhibiting several promising marketing behaviors.
+
+Segment builds this Audience with the `Page Viewed` event and Likelihood to Purchase and Predicted LTV Predictive Traits. Audience members have been active on your site within the last two weeks, have a high predicted lifetime value, and a high propensity to purchase. 
+
+#### Dormant
+
+Choose a **Dormant** Predictive Audience to target inactive customers. 
+
+Segment builds this Audience with the `Page Viewed` event and the Likelihood to Purchase Predictive Trait. Audience members have a low likelihood to purchase and haven't been active on your site in the last 60 days.