Product Recommendations in WhatsApp
Overview
Netcore Recommedation is an AI-powered engine that helps you deliver personalized product suggestions to users across channels such as WhatsApp, and Emails.
Recommendations use data from your website, such as product catalog, user behavior, and purchase history, to recommend relevant products to each user, increasing engagement, conversions, and revenue.
Product Recommendations in WhatsApp deliver dynamic, AI-powered product suggestions directly within WhatsApp campaigns. These recommendations are powered by Unbxd. The feature supports both Journey-based and Broadcast campaigns.
- Powered by the same AI mechanism used for Email and Web channels.
- Delivers personalized or mass-targeted recommendations depending on campaign type.
Prerequisites
Before you begin creating a recommendation-based email template, ensure the following conditions are met:
- Ensure your Netcore CE dashboard is mapped to the Unbxd Recommendations panel. The Netcore Support Team manages this integration.
- Obtain your Site Key and API ID from the Unbxd panel. Contact your CSM or Netcore Support to generate this.
- Complete the Unbxd JavaScript integration on your site.
- Access to Netcore Recommendations.
Refer here to know how to activate **Recommendations** for your account.
Important Points to Remember
- In a dynamic carousel, you only need to define the content for the first card. The media type (image), text and buttons, and will remain consistent across all cards, while the media and variables will be personalised for each card.
- If your template is not visible in the Campaign listing, it means the selected recommendation algorithm type is not supported.
- All recommendation algorithm types are supported for Journeys.
- If the UNBXD panel is not activated, Text and Media templates will not work. Only Carousel templates work for Recommendations.
- If Recommendations are disabled or inactive, a Dynamic template will behave like a Static template. In this case, we recommend creating a Static template directly.
- When creating a template, you only need to design one card. It will be automatically personalized for each user based on the selected recommendation algorithm.
Create WhatsApp Carousel Recommendation Template
Refer to this document to create a WhatsApp template.
Note
- If Recommendations are not enabled for your CE dashboard, selecting a Dynamic template will behave like a Static template.
- Use a Static template directly in such cases to avoid unexpected behavior.
- Select Recommendation from Display products by dropdown. Refer here to know more.
Refer to the given table to know the available options.
| Option / Field | Description |
|---|---|
| Websites | Select the relevant website. This determines which catalog and user data the recommendation engine will use. |
| Recommendation Type | Select an algorithm (e.g., "Recently Viewed", "Trending Products"). Example (Journey): Recently Viewed Example (Broadcast): Trending Products Note: Avoid using special characters (e.g., $, @, #) in algorithm names on Unbxd. These can cause delivery issues in WhatsApp messages. |
| Message Bubble | Enter the primary message text users will see. |
| Header Media | Only Image is supported in the header. |
| Media URL | Insert a dynamic media URL based on product properties. Learn more |
| Body Text | Add personalized content using product properties. Supports basic formatting (Bold, Italic, Strikethrough, emojis). |
| Buttons | Add up to two interactive buttons for CTAs like "View Product", "Buy Now". |
Important Points to Remember
While creating algorithms on the Unbxd panel, avoid using special characters like $, @, #, and so on, because using these can negatively impact email communications.
For example: Most Viewed Price 20 DASH 60$, the $ can lead to the WhatsApp message being dropped.
- Click SAVE AND SUBMIT. Review before sending for approval
- Give your template a suitable name, select the Marketing category and select Language from the dropdown.
- Preview your template on the right pane and then SEND FOR APPROVAL to Meta.
Predefined Recommendation Algorithms
Refer to the table below, which displays the predefined set of algorithms Netcore offers.
| Recommendation type | Description | Supported in Campaign | Personalised | Supported in Journey |
|---|---|---|---|---|
| 1. Best Selling products | Recommends top-selling products based on overall sales data. | Yes | Yes | |
| 2. Most Viewed products | Shows products with the highest views across users. | Yes | Yes | |
| 3. Top Sellers | Highlights top-performing products, often based on specific categories. | Yes | Yes | |
| 4. Trending products | Recommends products that are currently gaining popularity. | Yes | Yes | |
| 5. Recently Viewed | Shows products the specific user has recently interacted with. | No | Yes | Yes |
| 6. Frequently Bought Together | Recommends items that are often purchased together. | Yes | Yes | |
| 7. Similar Products | Suggests products similar in category or attributes to the one being viewed. | Yes | Yes | |
| 8. Cart-Specific Recommendations | Suggests products based on the current contents of the user’s cart. | No | Yes | |
| 9. Recommendation based on last purchase | Recommends products based on the user’s most recent purchase. | No | Yes | Yes |
| 10. Recommendation based on last viewed | Suggests products related to the last product the user viewed. | No | Yes | Yes |
Updated 6 days ago