Supplying the Recommendation-Engine with Sales Data
In addition to the product data already available through the Search function, the FACT-Finder Recommendation Engine uses sales information that you will also need to supply to us.
There are two basic options available for doing this. You can either send the data through the Tracking interface, or you can make it available in a file that you submit at specific intervals.
Tracking
This is the recommended method, since tracking integration also offers other benefits, such as additional analytical options. The Tracking interface is also recommended for FACT-Finder standard integration, so there should be no additional effort required.
A more detailed description of the features offered by the Tracking interface and how to use them can be found in the Integration document. In order to provide sales data for use by the Recommendation Engine, tracking must be integrated through to the buy event.
The big disadvantage of this method is that sales are only sent to FACT-Finder from the point of integration. If the Tracking interface is only integrated for the Recommendation Engine or this is an initial setup, there is a brief learning period. If you want to avoid this, you can provide previous sales information in an export file that FACT-Finder can also use as a source for importing recommendation data.
The tracking data must first be prepared by Analytics before it can be used by the Recommendation-Engine. The WhatsHot job then collects this data periodically (usually once a day) from Analytics and stores it in the APP_RESOURCES/analytics directory. The Recommendation-Engine import only works if there is data available.
Export file
The same principles that apply to exporting product data also apply when creating this export file. For more information on these principles, see: Record structure for export/import.
A CSV text file is the ideal format for this data. It is important to note that the export must be a complete export – incremental deliveries are not supported. If this results in a very large volume of data, a practical solution is to limit the exported period to 6-12 months.
The file must contain the following information:
| Name | Description |
|---|---|
Timestamp | Date and time of purchase. |
Product-ID | ID or article number of the purchased product. This identifier must also be present in the product data provided to the FACT-Finder Search tool. In the product data, this must be the field with the field role productNumber. |
Quantity | Quantity of the product that has been purchased. |
Shopping Basket ID | The shopping basket ID is required to establish which products were purchased together. This ID must be unique per shopping cart. If it is missing or 0, the corresponding event is ignored. |
User ID | The user ID identifies the purchaser. If the user is logged in when making the purchase, the shop’s own user ID is used for this parameter. If your web shop supports purchases without logging in, this must be an ID that is unique to each purchase. |
If variants exist in the product data and therefore a masterArticleNumber exists, it is recommended to pass the masterId as well to the Recommendation-Engine, as well. When a variant of a product is purchased, the engine then learns the connection to the master product.
Example
Timestamp;ProductID;Amount;CartID;UserID
2009-03-26 00:08:10;16041987;1;3880;23
2009-03-26 00:08:10;4582657;1;3880;23
2009-03-26 00:08:10;8954245;1;3880;23
2009-03-26 00:28:25;5659536;2;3881;30
2009-03-26 00:08:10;4582657;1;3882;42
2009-03-26 00:28:25;4571231;1;3883;51
...