Taking control over recommendations
      Back to home
      1. The Aqurate knowledge base
      2. Guides and How-to's
      3. Taking control over recommendations

      Fine-tuning recommendations

      Find out how to change the calculation method for recommendation types that support fine-tuning at API level.

      ⏳ Duration: 10 minutes 
      🎚️ Complexity: high

      📌 Prerequisites

      1. An Aqurate account (get it at aqurate.ai)

      2. An active integration of your Shop data (connect your shop here)

      3. An active subscription with Aqurate Personalize

      📖 Resources

      Check out and test our endpoints in the Recommendation section of our docs.

      Recommendation types that can be fine-tuned

      To give you more flexibility, some types of recommendations support additional settings, so you can fine-tune how recommendations are calculated. The types of recommendations that can be fine-tuned are Bestsellers (at store, category, or item level) and Recommended for you. All changes are done by using the specific API calls described below.

      You can also take control over what recommendations are shown in what context in order to prioritise or deprioritise specific products by using the Merchandising feature.

      Bestsellers

      1. The Popular now (/recs/bestseller) recommendations are affected by the following:

      • /recs/bestseller/set-lookback-window/{lookback_window} is used in order to select over how many days the bestsellers should be calculated. Default is 14.
      • /recs/bestseller/set-method/{method} is used to determine whether bestsellers are computed based on value, volume, or balanced. Default is volume.

      2. The Popular in category (/recs/item/bestseller-in-same-category) recommendations are affected by the following:

      • /recs/item/bestseller-in-same-category/set-lookback-window/{lookback_window} is used in order to select over how many days the bestsellers should be calculated. Default is 14.
      • /recs/item/bestseller-in-same-category/set-method/{method} is used to determine whether bestsellers are computed based on value, volume, or balanced. Default is volume.

      3. The Popular in same category (/recs/category/bestseller) recommendations are affected by the following:

      • /recs/category/bestseller/set-lookback-window/{lookback_window} is used in order to select over how many days the bestsellers should be calculated. Default is 14.
      • /recs/category/bestseller/set-method/{method} is used to determine whether bestsellers are computed based on value, volume, or balanced. Default is volume.

      Recommended for you

      The Recommended for you (/recs/user) recommendations are affected by the following:

      • /recs/user/set-max-orders/{max_orders} is used to exclude products that have been already purchased by the user. Default is 0.
      • /recs/user/set-max-views/{max_views} is used to exclude products that have already been viewed by the user. Default is 0.

      You’re done 🎉!

       

      Questions about using our API? We're excited to help: support@aqurate.ai