Overview:

The Upsell widget is a dynamic checkout block that shows targeted products, optional discounts, and conditional messages, designed to increase average order value without disrupting the checkout flow.

1. Product source options

The Upsell widget supports multiple product source modes so you can control exactly which products appear:

  • Manual: Select specific products or collections that always show in this widget (for example “Gift Wrap”, “Extended Warranty”, “Bestseller Add‑ons”).
  • Automatic (Shopify Search & Discovery): Let Shopify’s recommendation engine return “Related” or “Complementary” products based on the current cart and your Search & Discovery configuration.
Example – Manual vs Automatic
  • Manual: “Always show our ‘Gift Box’ product when any jewelry item is in the cart.”
  • Automatic: “For any product in the cart, show complementary items (e.g., chains, cleaning kit) configured in the Search & Discovery app, so you don’t need to update the widget when you add new SKUs.”

When using Automatic mode powered by Shopify Search & Discovery, you can choose between two behaviors:

  • Related: Surfaces alternatives and similar items (same category, style, or price range), useful for upgrades or side‑grade suggestions (e.g., “Premium Jeans” instead of standard jeans).
  • Complementary: Shows products that pair well with the current cart item (e.g., belt for pants, case for phone, filter for camera), ideal for true cross‑sell.
Example – Complementary recommendations

Cart: “Coffee Machine”

  • Complementary upsell might show “Filter Pack”, “Milk Frother”, “Coffee Beans” configured in Search & Discovery as complementary products.
  • When the merchant adds a new filter variant or a new coffee bean SKU, it can automatically participate in the upsell without changing the widget configuration.

3. Discount integration and auto‑apply

The Upsell widget can be linked to a discount code so the offer is applied automatically when the customer adds a product from this block.

  • When a shopper clicks “Add” on an upsell product, the widget adds the item and immediately attaches the configured discount code to the checkout, without requiring the customer to type the code.
  • You can then write titles and descriptions that clearly advertise the offer, such as “Add this and save 10%” or “Free gift when you add this item”, while the discount logic is fully handled by the app’s discount engine.
Example – Free gift for subscription

Business rule: “If a subscription item is purchased, show a free gift in upsell and auto‑apply discount.”

  • Configure Upsell widget with a free‑gift product source and attach a “100% off” product discount code that only applies when the upsell item is added.
  • Customer experience: As soon as they add the free gift from the widget, the discount code is silently applied, and the order summary shows the gift at ₹0 or the discounted price.

4. Layout types: List and Grid

You can control how upsell products are visually presented using two layout modes:

  • List layout: Vertical list with more space for text, ideal for 1–3 high‑intent offers where you want to explain benefits (e.g., warranty, protection plan, subscription upgrade).
  • Grid layout: Multi‑column grid that shows several products at once, better for “You may also like” style recommendations (e.g., 4 small accessories or multiple flavor options).
Example – When to use which
  • High‑value single offer: “Add 2‑year protection plan” in List view with icon, short description, and one CTA.
  • Multi‑option: “Complete the look” grid with 4 accessories, each with image, price, and quick add button, allowing the customer to add more than one upsell item.

5. Conditional display rules (Triggers)

Like the Discount Conditions system, the Upsell widget can be shown only when specific rules are satisfied, so offers remain relevant and on‑brand.

  • Triggers can check cart contents (products, collections, quantity, subtotal), customer attributes (tags, segments), or purchase type (subscription vs one‑time) before rendering the widget.
Example – Free gift only for subscription

Business requirement: “Show a free claim gift widget only if a subscription item is purchased.”

  • Condition group:
  • Cart contains product(s) from “Subscription” collection OR line items marked as subscription.
  • Optional: Subtotal ≥ ₹1 to avoid empty‑cart edge cases.
  • Result:
  • Subscription in cart → Upsell widget appears with the free gift offer.
  • Only one‑time items in cart → Widget stays hidden, avoiding misleading promotions.

6. Upcoming AI‑based recommendations

An upcoming AI mode will extend Automatic recommendations with a dedicated AI engine focused on checkout‑stage personalization.

  • The engine will combine multiple signals such as cart contents, browsing history, purchase history, product attributes, and popularity trends to predict which product has the highest probability of being added at that moment.
  • Instead of a fixed list, each customer can see a different product set, ranked by an ML model (e.g., collaborative filtering + deep ranking) that is continuously retrained on conversions and rejected offers.
Why AI recommendations convert higher
  • AI engines typically outperform manual or static rules by surfacing items that are both context‑aware (match current cart) and person‑specific (match user’s past behavior), which is shown to increase click‑through, add‑to‑cart, and conversion rates.
  • At checkout, where intent is already high, these highly targeted AI suggestions can achieve multiple‑times higher purchase likelihood than a fixed carousel, especially on returning customers and high‑traffic stores; in this app, the goal is around 3× improvement versus non‑personalized upsell blocks.
Example – AI vs fixed list

Fixed mode:

  • Always shows “Gift Wrap”, “Bestseller Tee”, “Sticker Pack” to everyone.

AI mode:

  • For someone with sports gear in cart and a history of buying caps, AI recommends “Team Cap” first.
  • For a first‑time visitor buying a gift set, AI prioritizes “Gift Wrap” and “Express Shipping Upgrade”.
  • The system learns from which upsells get accepted or ignored and automatically re‑ranks future suggestions to maximize revenue.