Predictions
Easily predict future customer behavior to deliver precise and personalized campaigns without needing to tap data science teams.
We'll get back to you shortly. For now, you can create your workspace by clicking below.
Predictions gives marketers the power of predictive AI, making it easy to uncover patterns in historical customer data, predict future behavior, and discover audiences primed for engagement—saving teams time, boosting campaign performance, and improving personalization.
“With Predictions, our marketing team can now easily predict which users are most likely to bid on an item being auctioned on our marketplace, make a purchase, or even create a listing for a specific product category, then run targeted campaigns across channels—without needing additional data science resources.”
Predictions gives marketers easy-to-use, out-of-the-box predictive models built on trusted, AI-ready data in Segment to quickly create Predictive Traits that get added to the customer profile to power more timely, targeted, and precisely personalized customer experiences on every channel.
Likelihood to Purchase
Predict the propensity of customers to make a purchase.
Likelihood to Churn
Proactively identify customers who are likely to stop using your product.
Predicted Lifetime Value (LTV)
Predict a customer's future spend over the next 90 days.
Custom Predictive Goals
Predict the propensity of customers to perform any event tracked in Segment (e.g. likelihood to use a promo code or refer a friend).
Predictions includes pre-built Predictive Audiences to inspire new campaign ideas and help you quickly create AI-powered audiences in just a few clicks.
Ready to buy audiences
Includes customers who are most likely to purchase.
Long shots audiences
Includes customers who have previously interacted with your brand but aren't currently engaged.
High LTV audiences
Includes customers with high predictive lifetime value.
Potential VIPs audiences
Includes recently active customers with high predictive lifetime value as well as a high likelihood to purchase.
Dormant audiences
Includes inactive customers who are unlikely to purchase.
Easily identify, target, and launch campaigns for users who are most likely to convert or take a desired action across all your customer channels and journeys.
Improve ad targeting and return on ad spend (ROAS)
Build lookalike audiences based on customers with a high likelihood to purchase or high LTV. Suppress ads to users with low likelihood to purchase or low LTV to conserve ad spend.
Increase performance of promotional and lifecycle campaigns
Create more personalized emails, SMS, WhatsApp, and push notifications for customers no matter where they are in their journey or which buying groups they are in—like discount vs full-price buyers.
Personalize online experiences
Update your web and mobile experiences to showcase different products or promotions based on their propensity to purchase or engage with your content.
Boost upsell and cross-sell
Drive greater upsell and cross-sell opportunities by launching timely campaigns showcasing personalized items when customers are most likely to buy.
Enhance loyalty programs
Quickly spot customers with high predicted LTV to invite into VIP loyalty programs or send special personalized offers.
Bolster win-back campaigns
Proactively identify disengaged or at-risk customers for targeted win-back and retention campaigns.
Predictions provides transparency into every prediction you create, so your marketing and data teams always feel confident in your predictions.
The Explore view visualizes the prediction data and lets marketers easily cohort that data to create audiences for targeted campaigns or customer journeys.
The Understanding view offers insights into all the underlying data points and how they contribute to the model.
“By using Predictive Audiences, we have significantly improved our paid search performance — achieving a 124% higher click-through rate (CTR) and 102% higher conversion rate (CVR) than our average search engine marketing (SEM) performance.”
Start connecting your data with Segment.