Recommendation Strip

The Recommendation Strip is a product recommendation widget that displays dynamically generated product suggestions based on your store's order history. Unlike traditional recommendation systems that require complex integrations, the Recommendation Strip uses your existing purchase data to identify patterns and surface relevant products to your customers.

Overview

The Recommendation Strip appears as a horizontal or vertical carousel of product thumbnails. It can be positioned as an overlay on your page or embedded inline within a specific container. The widget automatically determines which products to recommend based on the page the customer is viewing and your store's purchase patterns.

Order Data Source

The Recommendation Strip requires an order data source rather than the traditional text-based data sources used by other agents. Order data contains structured JSON representing your store's purchase history, enabling the system to analyze buying patterns.

Data Requirements

• Minimum Volume: We recommend at least 1,000 orders for accurate pattern detection. Fewer orders may result in less relevant recommendations.
• History Depth: Include at least 3 months of order history to capture seasonal patterns and product relationships.
• Product Details: Each product in your order data should include a display name, product URL, category, SKU, and thumbnail URL.

How Recommendations Work

The Recommendation Strip uses a smart, multi-strategy approach to generate relevant product suggestions. The system automatically detects the type of page a customer is viewing and applies the most appropriate recommendation strategy.

Product Pages

When a customer views a specific product page, the system analyzes your order history to find products that are frequently purchased together with the viewed product. This "customers who bought this also bought" approach surfaces complementary products and accessories that real customers have paired with the current product.

Category Pages

On category or collection pages, the system identifies products from other categories that customers frequently buy alongside items from the current category. This cross-category approach helps customers discover related products they might not have considered, like suggesting accessories when browsing electronics or matching items when viewing apparel.

General Pages

For homepage, cart, or other non-product pages, the system displays your store's trending products — items with the highest purchase frequency across recent orders. This ensures customers always see popular, proven products even when there's no specific context to personalize recommendations.

Content Customization

The Recommendation Strip offers extensive options for controlling what products appear and how the widget behaves on your pages.

Display Settings

• Strip Title: Customize the heading displayed above the product carousel (e.g., "You May Also Like" or "Recommended For You").
• Display Position: Choose how the strip appears on your page:
  • Inset: Embed the strip within a specific HTML element by ID
  • Bottom Overlay: Horizontal strip fixed to the bottom of the viewport
  • Left/Right Overlay: Vertical strip fixed to the side of the viewport
• Inset Container ID: When using Inset position, specify the HTML element ID where the strip should be rendered.
• Trigger Delay: Set a delay (in seconds) before overlay positions appear. Useful for letting customers engage with content before showing recommendations.

Product Limits

• Maximum Products: Limit how many products appear in the carousel (1–12 products).
• Minimum Products: Set a threshold below which the strip won't display. If the system can't find enough relevant products, the widget stays hidden rather than showing sparse recommendations.

Exclusions

Control which products never appear in recommendations:

• Excluded Categories: Remove entire product categories from recommendations (e.g., exclude "Clearance" or "Discontinued" categories).
• Excluded Product Codes: Block specific parent products from appearing.
• Excluded SKUs: Block specific product variants from appearing.

Default Products

Configure fallback products for scenarios when AI recommendations aren't available or you want more control:

• Default Products List: Manually specify up to 10 products as fallbacks. These appear when the system can't generate AI recommendations.
• Only Show Defaults: Skip AI recommendations entirely and display only your curated default products. Useful for promotional campaigns or featured collections.
• Always Display Defaults: Append your default products to AI recommendations, ensuring specific products always appear.
• Fill to Maximum: When AI returns fewer products than your maximum, automatically fill remaining slots with default products.

Styling

The color and overall style of the Recommendation Strip can be customized using the intuitive WYSIWYG tool in your dashboard. The styling options adapt based on your selected display position.

Theme & Typography

• Color Theme: Choose between light and dark base themes.
• Font Family: Match your site's typography with Sans Serif, Serif, or a custom font family.
• Title Styling: Customize the strip title color and font size.
• Product Name Styling: Adjust the color and size of product names beneath thumbnails.

Container

• Background Color: Set the strip's background.
• Border: Toggle borders on/off and customize border color, size, and radius.
• Close Button Color: Style the dismiss button for overlay positions.
• Thumbnail Size: Control product image dimensions (80–200px).
• Thumbnail Border Radius: Round the corners of product images.

Overlay Size

When using overlay positions, you can control how much screen space the strip occupies:

• Horizontal Width: For bottom overlays, set the width as a percentage of the viewport (50–100%).
• Vertical Height: For left/right overlays, set the height as a percentage of the viewport (50–100%).

Analytics

BadgerFy.ai tracks various metrics related to your Recommendation Strip's performance, including strip displays, product clicks, and conversion events. Use the analytics dashboard to understand which products are being recommended and how customers interact with the widget.