> For the complete documentation index, see [llms.txt](https://docs.codegiganten.de/plugin-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.codegiganten.de/plugin-documentation/en/apps/cross-selling-pro-ai-cart-cross-selling-bundles-and-sets/recommendation-types.md).

# Recommendation types

All recommendation types of Cross Selling Pro: what they show, which data they need, what happens without data and from which plan they are available.

The recommendation type decides which kind of products a template shows: similar products, bestsellers, recently viewed products and so on. You choose it in the "Recommendation type" field of a recommendation template. This page shows what each type is for, which data it needs and what happens when no data is available yet.

In new shops it is worth checking the data requirements: some types work from day one, others only once orders and visitor behaviour are available.

## Overview

| Recommendation type in the Administration                                | Shows                                                    | Needs                                                       | Without data                      | From plan |
| ------------------------------------------------------------------------ | -------------------------------------------------------- | ----------------------------------------------------------- | --------------------------------- | --------- |
| "Similar products (alternatives)"                                        | Products that match the current product                  | A source product and your product data                      | Calculates with product data only | Basic     |
| "Recently viewed"                                                        | Products the visitor has viewed recently                 | Visitor identifier, product views of the last 30 days       | No result                         | Basic     |
| "Bestsellers"                                                            | The most ordered products                                | Orders                                                      | No result                         | Starter   |
| "Buy again (products from your order history)"                           | Products the customer has ordered before                 | Customer reference, own orders                              | No result                         | Starter   |
| "Trending now"                                                           | Products that currently get a lot of attention           | Product views, with insight analysis also clicks and orders | No result                         | Expert    |
| "You may also like"                                                      | Suggestions based on the visitor's previous behaviour    | Visitor identifier, visitor behaviour                       | No result                         | Expert    |
| "Goes well with (frequently bought together)"                            | Products often ordered together with the current product | Many orders                                                 | Most ordered products             | Expert    |
| "Continue where you left off (products added to cart but never ordered)" | Products that were in the cart but never ordered         | Visitor identifier, add-to-cart actions                     | No result                         | Expert    |
| "Inspired by your wishlist"                                              | Products that match the wishlist                         | Products on the wishlist                                    | No result                         | Expert    |

"No result" means: InsightSearch returns an empty list. Then the fallback you set in the "Display" section of the template applies ("Fallback type", for example "Other template"). Without a fallback the block stays empty.

{% hint style="info" %}
Locked types appear in the selection list with a lock and the plan name. If you choose a type that is not included in your plan, a notice with "Compare plans" appears. More in [Requirements and plans](https://docs.codegiganten.de/plugin-documentation/en/apps/cross-selling-pro-ai-cart-cross-selling-bundles-and-sets/requirements-and-plans.md).
{% endhint %}

## Good to know first

* **Source product:** "Similar products" always needs a product to recommend for. On the product page this is the product shown, in the cart a line item of your choice.
* **Visitor identifier:** "Recently viewed", "You may also like" and "Continue where you left off" need an identifier for the visitor. Without it nothing appears. How the identifier is created and which consent it needs is described in [Tracking and data protection](https://docs.codegiganten.de/plugin-documentation/en/apps/cross-selling-pro-ai-cart-cross-selling-bundles-and-sets/tracking-and-data-protection.md).
* **Cancelled orders** never count for Bestsellers, Goes well with and Buy again.
* **Weighting and boosters** only apply to "Similar products". For all other types the order comes from behaviour data. See [How recommendations work](https://docs.codegiganten.de/plugin-documentation/en/apps/cross-selling-pro-ai-cart-cross-selling-bundles-and-sets/how-recommendations-work.md).

## The types in detail

### Similar products (alternatives)

The default type for new templates and the most versatile one. InsightSearch compares the source product with your range: brand, category, price, attributes, AI similarity and, once available, the shopper behaviour of your visitors. The note in the Administration sums it up: it works from day one, even without orders.

If a product has no AI data or shopper behaviour yet, InsightSearch continues with the remaining data. The product is not ranked lower because of that.

**Typical use:** alternatives on the product page, accessories in the cart (with the "Cart: matching accessories" preset), recommendations in the off-canvas cart.

{% hint style="info" %}
Without shopper behaviour this type tends to show similar products rather than accessories. For accessories, rules help (for example a category other than the source product's) or the "Accessories (product page)" preset.
{% endhint %}

### Recently viewed

Shows the product pages the visitor has opened in the last 30 days, newest first. For logged-in customers, views from earlier visits are taken into account as well.

**Typical use:** home page, customer account, category pages. A new visitor without views sees nothing, so set a fallback.

### Bestsellers

The most ordered products of your shop. InsightSearch first looks at the last 30 days. If that is not enough, at the last 90 days, then at all available orders. It counts in how many orders a product appears, not the quantity.

With the insight analysis switched on (section **Products & data → AI**), InsightSearch also uses a bestseller list calculated daily in advance.

**New shops:** orders of the last 90 days can be transferred afterwards in **Products & data → Sources** with "Sync" as soon as the plugin finds orders that have not been transferred yet. This gives the type data from the start.

### Buy again (products from your order history)

Shows products the customer has ordered before. The most frequently ordered products come first, with equal counts the most recently ordered. Useful for consumables such as tea, pet food or printer cartridges.

This type needs the customer reference. "Send customer reference" must be active in the plugin configuration. Guests and new customers see nothing.

**Typical use:** customer account, home page for logged-in customers.

### Trending now

Products that currently get a lot of attention. How trends are calculated depends on the insight analysis:

* **With insight analysis:** products that were in clearly higher demand in the last 7 days than in the 7 days before. The basis is clicks on search results, orders and impressions in the search.
* **Without insight analysis or with too little data:** the most viewed product pages of the last 14 days.

If you do not use the InsightSearch search plugin, there is hardly any search data. The list then mainly relies on product views.

### You may also like

Suggestions based on the visitor's previous behaviour. InsightSearch keeps a profile per visitor or customer. Views, clicks, add-to-cart actions and purchases have different weights: a purchase counts much more than a view. For logged-in customers the customer profile applies, otherwise the profile of the visit. Profiles without activity for 30 days expire.

**Typical use:** home page, customer account. New visitors without behaviour see nothing, so set a fallback.

### Goes well with (frequently bought together)

Products that appear in the same order as the current product more often than average. A pair only counts from three shared orders. In expert mode you can change this limit with "Minimum co-purchase count". InsightSearch first looks at 30 days, then 90 days, then all orders.

The note in the Administration says that the type uses real order data and needs an order history, so new shops see only a few results at first. If pairs are missing, InsightSearch fills up with the most ordered products of your shop. Without a source product the result is essentially a bestseller list.

**Typical use:** product page below the buy button, cart.

### Continue where you left off

Products the visitor added to the cart but never ordered, newest first. InsightSearch first looks at 30 days, then 90 days, then all data.

**Typical use:** home page and customer account for returning visitors.

### Inspired by your wishlist

Takes up to six products from the customer's wishlist and looks for similar products. The wishlist products themselves do not appear. Without a wishlist the result is empty.

**Typical use:** customer account, home page for logged-in customers.

## Types outside the templates

Two further kinds of recommendations are not selected in the "Recommendation type" field. They run in fixed places:

* **Milestone recommendations:** products whose price closes the gap to the next milestone of the progress bar, for example up to free shipping. From Starter, matching products from the price range are chosen; from Expert, similarity is taken into account as well. See [Milestone recommendations](https://docs.codegiganten.de/plugin-documentation/en/apps/cross-selling-pro-ai-cart-cross-selling-bundles-and-sets/milestone-recommendations.md).
* **Recommendations for a search term:** if you also use the InsightSearch search plugin, it shows fallback products for searches without results and "Goes with …" in the search overlay. From the Starter plan. See [Working with InsightSearch, cogi Commander and other plugins](https://docs.codegiganten.de/plugin-documentation/en/apps/cross-selling-pro-ai-cart-cross-selling-bundles-and-sets/working-with-other-plugins.md).

## Which type for which goal?

| Goal                                     | Suitable type                                                                       |
| ---------------------------------------- | ----------------------------------------------------------------------------------- |
| Alternatives to the product being viewed | Similar products                                                                    |
| Accessories and add-ons                  | Similar products with the "Accessories (product page)" preset, later Goes well with |
| Show new visitors something safe         | Bestsellers                                                                         |
| Bring returning visitors back            | Recently viewed, Continue where you left off                                        |
| Serve regular customers                  | Buy again, You may also like, Inspired by your wishlist                             |
| Show current demand                      | Trending now                                                                        |

For types that need visitor data, a template with "Bestsellers" or "Similar products" works well as a fallback.

## Read more

{% content-ref url="/pages/A5tmmG01QN0sNwbBOlS6" %}
[How recommendations work](https://docs.codegiganten.de/plugin-documentation/en/apps/cross-selling-pro-ai-cart-cross-selling-bundles-and-sets/how-recommendations-work.md)
{% endcontent-ref %}

{% content-ref url="/pages/hSRY2MxmbdY5dCCa9KY4" %}
[Creating recommendation templates](https://docs.codegiganten.de/plugin-documentation/en/apps/cross-selling-pro-ai-cart-cross-selling-bundles-and-sets/recommendation-templates.md)
{% endcontent-ref %}

{% content-ref url="/pages/H7hG7KNHBotOdMhNijf5" %}
[Requirements and plans](https://docs.codegiganten.de/plugin-documentation/en/apps/cross-selling-pro-ai-cart-cross-selling-bundles-and-sets/requirements-and-plans.md)
{% endcontent-ref %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.codegiganten.de/plugin-documentation/en/apps/cross-selling-pro-ai-cart-cross-selling-bundles-and-sets/recommendation-types.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
