Marketers and e-commerce managers now have the option to tune and optimize their Raptor solution to fit their specific needs and requirements.
We have added a visual output on the module page, which enables Raptor users to test the individual module within the interface of the control panel.
The new module page provides an opportunity to debug and evaluate the output of a module before any changes are deployed.
1. It is now possible to define the type of behavioral data that the algorithm applies when presenting relevant recommendations.
2. Users can now specify how personal their recommendations are. We have added an option to define the weight of different contextual dimensions, by adjusting the balance between browsing and buying behavior in a module’s output.
For example, this option is useful in a grocery shopping scenario. If two users are browsing canned tomatoes, one might need pasta and parmesan, and the other is looking for salsa and taco shells.
In this case, the recommendations need to be highly personalized in order to detect the context of individual users and provide them with relevant recommendations that match their intent.
3. Furthermore, we have included a formatting option, which enables users to freely choose the format of our data payload. E.g. Json, XML, HTML, or even images.
4. Finally, we have added a filtering option to the module page, making it possible to exclude items in a recommendation set. Users can also choose to exclude products that do not live up to a certain value or property, such as size, color, stock status, margins and more.