Skip to main content

Predictions

Overview

Skypoint prediction model provides multiple options that leverage AI and machine learning to assess historical data, discover patterns, observe trends, and use that information to predict future trends. The prediction options are available from building basic yes/no or true/false predictions using AI builder to handling intricate scenarios using custom machine learning models that are tailored to your needs. Skypoint predictive insights allow you to anticipate customer needs with real-time data and deliver meaningful experiences across the entire customer journey.

Skypoint Lakehouse is integrated with MLflow to provide a powerful combination for managing and deploying machine learning models. You can store and manage your data and models in the same place. Important features of Skypoint-powered prediction models are as follows:

  • Increase revenue - Leverage real-time data to determine who is most likely to buy, which segment will churn, or when to present an offer.
  • Support teams - Collect valuable BI and predictive insights for marketing, sales, service, and IT through Skypoint Lakehouse.
  • AI-powered insights - Predict customer needs by unlocking powerful insights using pre-built AI and ML models or custom models.
  • Analyze and adapt - Learn from patterns in historical customer data to identify risks and opportunities and make predictions about future outcomes.

[!TIP] Monitor and update the prediction model and audience segmentation to stay updated with any changes in customer behavior or characteristics. You can configure the data update schedule during configuration.

Skypoint Predict

Skypoint Predict section allows you to access predictions and custom models. This section includes two options:

Prerequisite

  • You must have completed Master Data Management (MDM) processes.
  • Topic modeling and sentiment analysis is an exception and does not need MDM configuration.

Skypoint Predict

Skypoint Predict section allows you to access predictions and custom models. This section includes two options:

  • Predictions: View any existing predictions and create new ones.
  • Custom models: View any existing custom models and import new ones.

Alt image

Predict customer needs with machine learning and AI

There are multiple out-of-box prediction models available on Skypoint Studio. These are as follows:

  • Customer churn model - Predict which customers are likely to cancel or stop using a company's product or service.
    • Transactional churn - Predict which customers are likely to stop making individual purchases or transactions with a company.
    • Subscription churn - Predict whether a customer is at risk for no longer using your company’s subscription products or services.
  • Product recommendation model - Predict and recommend products to customers based on their purchase history and patterns, as well as the behavior of similar customers.
  • Customer Lifetime Value (CLV) model - Predict the potential revenue that individual active customers will bring to your business.
  • Sentiment analysis & topic modeling – Predict comprehensive sentiment analysis (sentiment model and topic modeling) to get the most actionable insights from feedback, reviews, and surveys.
  • RFM model - Use recency, frequency, and monetary value to predict which customers are most likely to respond to a specific offer or promotion.

Custom models

Integrate your artificial intelligence and machine learning models deployed as web service endpoints to utilize unified tables (e.g., predictions on unified customer profiles and activities) and perform a specific task with a high degree of accuracy.

Using an API, you can submit single or bulk input and receive a corresponding prediction.

  • Single upload – When the model uses a single data sample, such as a single customer's purchase history.
  • Bulk upload – When the model uses a large dataset, such as a large collection of customer purchase histories.

Create a quick audience

Audience creation from the prediction models empowers you to quickly identify and segment users into homogeneous groups and target them with differentiated and personalized marketing strategies.

With built-in AI, Skypoint Cloud creates new dimensions for understanding and segmenting your customers into high-performing audiences.