RX Forecast™

RetentionX leverages advanced artificial intelligence and machine learning models to analyze historical data, uncover complex patterns, and generate high-confidence forecasts for future performance.

How RX Forecast™ Works

Businesses today use a wide range of tools—from basic spreadsheets to advanced financial planning software—to forecast future business outcomes such as product demand, financing needs, or overall financial health. Most of these tools rely on time series data, assuming that future performance can be derived from past trends. However, this approach can struggle to produce accurate forecasts, especially when working with large data sets or irregular patterns.

RX Forecast™ uses machine learning to go beyond simple trend extrapolation. It combines time series data with additional influencing variables to generate more robust forecasts. To do this, RetentionX analyzes historical performance alongside relevant related data that may impact future outcomes. For example, demand for specific product categories can vary based on seasonality or store location.

To give you an idea, RX Forecast™ may take the following related factors into account:

• Weekdays and holidays

• Economic trends

• Historical development of your KPIs

Identifying and weighting these correlations manually is complex. With the help of machine learning, RetentionX can detect and model these relationships at scale.

Where to Find RX Forecasts?

RX Forecast™ is available for the following time series reports within RetentionX:

• Total Sales
• Average Order Value
• Customer Development
• Repeat Customers

To view a forecast, simply unlock predictions using the forecast functionality in the respective report.

Based on your custom-trained model, RetentionX calculates three forecast scenarios to help you understand the full range of possible outcomes. This scenario-based approach allows you to plan with greater confidence by accounting for both upside potential and downside risk.

• Optimistic: Best possible forecast on the future data development

• Neutral: Average expected outcome across all forecasts

• Pessimistic: Worst-case forecast for future performance

RetentionX always applies a 3:1 data ratio for forecasting. This means that for every 3 units of historical data, RetentionX forecasts 1 unit of future time. This ratio ensures sufficient data is available to train reliable predictive models. For example, if you have three years of historical data, RetentionX will generate forecasts for the upcoming one year.

Not using RetentionX yet?