Model features are data points Convert's machine learning technology uses to generate its rankings. A model feature can be a product metric or attribute, such as buy-to-detail rate or color.
Convert will determine how closely correlated a given feature is to purchase behavior, and then weigh the feature accordingly when ranking a set of products.
To add sophistication to your ranking models beyond what's available from Google Analytics, you can create your own model features based on custom data you've imported.
For more information on importing data for use in model features, see our guide here.
Table of Contents
- Understanding Model Features
- Creating Custom Model Features
- Adding Custom Model Features to Ranking Models
Understanding Model Features
As mentioned above, a model feature can be any product attribute or metric that may have a relationship to the likelihood that the product will be purchased.
However, please note that creating a new model feature isn't as simple as uploading a spreadsheet with the color of each product in your catalog. To be useful to Convert's ranking models, a feature must be transformed into a specific numerical format. This complex process will require a background in data science or machine learning to be effective. For assistance, or more information, please contact Convert Support.
After transforming your data into a format useful to Convert, you'll need to upload it to a dataset. For instructions on importing data, click here.
Once you've imported your data, you can begin creating custom model features.
Creating Custom Model Features
Model features are created at the account level so that they can be used by ranking models within any site.
- Navigate to your Accounts list.
- Click on the Modeling button for the account.
- Click Add Feature.
- Enter a unique, identifiable Label.
- Use the dropdown to select the appropriate Repository.
- Use the dropdown to select the Dataset that contains your feature data.
- Enter the exact Field Name from your dataset to identify the data to be used.
To reduce potential errors, field names are not case sensitive. However, if you're uncertain what column header is used in the dataset, we recommend exporting it to confirm.
- Click Save.
Once the feature has been created, you can always change the repository, dataset, and field name configured.
Adding Custom Model Features to Ranking Models
For instructions on adding custom model features to a ranking model, view our guide on Managing Ranking Models.