RetentionX uses artificial intelligence, which includes machine learning, to leverage algorithms in order to learn from data to identify patterns and make forecasts and predictions.
While Forecast works for time series, RX Prediction™ is used to project metrics and actions for individual entities on a daily basis. RX Predictions answer the following questions:
- How much is this customer worth in his life cycle?
- How healthy is your customer base?
- What is the probability of losing this customer?
- Is this item likely to be returned?
To unlock this feature, please activate the Forecast Plus add-on.
How RX Prediction™ Works
RX Prediction™ – wait, what?
Prediction is performed via machine learning and searches data twins within historical and current data sets. Traditionally, you have only two options when developing machine learning (ML) models.
One option is to manually prepare the characteristics, select the algorithm, optimize the model parameters to have full control over the model design, and track all the thoughts that went into creating the model. However, this approach requires extensive ML expertise.
If one lacks this expertise, you can resort to an automated approach (AutoML) for model generation as a second option. This does all the work for you. A model created with AutoML can work well, but it can be less confidence-inspiring because you cannot track what went into the model. Also, you cannot recreate it or learn best practices that might be helpful in the future.
RX Prediction™ eliminates this choice. It lets you automatically apply machine learning models - without compromise. RX Prediction™ automatically explores different solutions to find the best model for your data.
Please keep in mind, that RX needs at least 1 month of history and 100 order lines to apply and train machine learning models.