Help Center

To create personalized product recommendations in BlueConic dialogues, you start by collecting product data using the BlueConic Product Collector Connection. This connection collects data about your products and stores it in a BlueConic product store to be used for personalization in BlueConic.

Make smarter personalized product recommendations with BlueConic

You can use this information to make smarter product recommendations in dialogues. Deliver one-to-one product recommendations using toolbar plugins that use filters and product recommendation algorithms to choose the optimal products to recommend based on individual customers' behaviors, interests, preferences, and previous purchases.

Use personalization to target customers with abandoned shopping carts

By detecting the number of items in the customer's shopping cart and the last update, you can target customers who have abandoned shopping carts to complete their purchases. Please note that this is a relatively complex use case, and you can reach out to the BlueConic Customer Support team if you need assistance.

Note: The Product Collector needs to be active on the page that fires the `shoppingcart` or `order` event on your site, such as the shopping cart or order page (not just on the product pages).

Configure the Product Collector to collect items for recommendations

To create a Product Collector connection, select Connections from the main BlueConic menu and click the Add connection button. Search for 'product collector' and create the connection. When the Product Collector connection appears, provide a name at the top of the page.

Email notifications: In the metadata section you can choose whether to get email notifications when the connection runs or fails to run.

To set up the connection, select "Collect data from your channels" in the lefthand panel.

1. Select the channel(s) to collect product data from.
   Optionally, define URL rules to specify which areas of the sites to collect product data from.
2. Next, paste a product URL into the "Test URL" area, and click 'Test' to review the metadata that would be collected for your product:
   
   If the "Name", "Description", "Image", "URL," and "Categories" fields are not populated, the webpage may not surface product information in a supported format. BlueConic supports:
   
   • Schema.org Microdata or JSON-LD
     • Getting started with schema.org using Microdata
     • The complete beginner’s guide to Schema.org markup
     • Structured data with schema.org - the ultimate guide
   • Open Graph meta tags
   • Regular meta tags
   To add custom metadata fields, select Add data field. For example, if your products have a typical gender association, or are tagged based on associated brands, influencers, or sources, you can use this data in recommendations placement filtering.
3. Some algorithms are based on a look-back time frame, for example the "top products" algorithm.
   You can configure this time frame based on hours or days here:
   

   

4. Optional: Add request headers, to be able to scrape content for webpages that are only available after login, after a paywall, or on development sites.
5. Detect products in the shopping cart, order data, and last time the cart was updated.
   A. Add shopping cart data to profile properties.
   Select profile properties to store the number of unique products in the customer's shopping cart. You can choose an existing profile property or create one here.
   
   B. Implement the JavaScript shopping cart event.
   Add the following JavaScript code to your website to let BlueConic know which products the customer adds to or removes from the cart, and when the cart is emptied. Use this data in dialogues for recommendations.
   

   // Replace the contents of the productIds array
   // with the IDs of all products that are currently in the shopping cart.
   
   blueConicClient.event.publish("shoppingcart", {
    productIds: ["id1", "id2", "id3"]
   });

   C. Implement the JavaScript order event.
   Add the following JavaScript code to your website to let BlueConic know which products the customer ordered. Use this data in dialogues to improve recommendations, using the "bought items" algorithm and collaborative filters and algorithms for order data.
   

   // Replace the contents of the productsIds array
       // with the IDs of all products the customer just ordered.
      
       blueConicClient.event.publish("order", {
       productIds: ["id1", "id2", "id3"]
       });

   Note: For both JavaScript options B and C, the product IDs in the event should exist in the content store. The configuration of the "ID" field should match the ID that is sent in the event.

   Learn more about BlueConic recommendation algorithms for personalization

   For details on the recommendation algorithms that power personalized product recommendations in BlueConic dialogues, see BlueConic recommendation algorithms.