Product grouping

The Solr product grouping feature enables grouping parent catalog entries with their underlying SKUs at query time. Products can be chosen to represent the group that is returned in search results to provide visual relevancy.

Product grouping can be useful for increasing search relevancy in your store. For example, use product grouping to expand the search scope of products and their underlying SKU, provide visual relevancy in search results, or promote a representative SKU in searches or store pages. For a detailed discussion of how product grouping works in WebSphere Commerce Search, see How product grouping works.

There are several ways to configure grouping, depending on your needs. The most common scenarios are described below. These scenarios link to the relevant configuration steps described in Enabling product grouping.

Scenario 1: Expand the search scope to include products and their underlying SKUs, and aggregate the results to return only the parent products

Most catalog entries are modeled to contain a product-SKU relationship. Each product can contain multiple SKUs. When using a searchType value that returns only products, the search does not include SKU-specific properties such as a SKU attribute value or SKU part number.

However, if another searchType value is used to return products and their underlying SKUs, or SKUs only, the search result contains many redundant results. These results include several SKU-specific properties that are too similar to each other. This makes it difficult to find relevant content in a search result that likely spans multiple search result pages.

For example, an apparel product might contain multiple items in different colors and sizes. Assuming the apparel is available in three colors and five sizes, there are 15 different possible combinations of items for that single product. If the search term is generic enough to generate hits on all 15 items, for example, dress pants, all 15 items are returned. This scenario might even be common among other apparel items, too, resulting in multiple search result pages.

Using product grouping prevents this scenario from occurring, since a searchType is used to search products and their underlying SKUs, but only the product is returned. This enables shoppers to search for SKU-specific properties, and result in a smaller search result set.


Scenario 2: Show facet values based on only entitled and published items, by rolling up SKU attributes and properties at run time, instead of index time

A method to generate search hits based on item attributes or properties is to roll up the attributes or properties to their parent's products at index time. By default, at index time, some of the SKU attribute values and properties are rolled up to their parent products to be used for searching and filtering. For example, this approach is used by default for the product's assigned Attributes Dictionary Attributes and inventory counts.

Using product grouping prevents the need to roll up SKU attributes and properties during index time, or sum up the inventory count of items. That is, the search scope includes both products and SKUs. As a result, shoppers are shown facet values based on their entitled contracts and published items.


Scenario 3: Show indexed price ranges for products in category pages

Using product grouping enables configuring the navigation pages to display accurate price ranges. The price ranges based on the indexed offer price of all the entitled published SKUs for a product.


Scenario 4: Provide visual relevancy for keyword searches

A product property can accept only one value. For example, only one thumbnail image can be assigned to a product, which might not match all the items thumbnails. When a specific item attribute or property is searched, or an item's attribute facet is selected (and a searchType value is configured to return products), the product's default property is loaded instead.

For example, clicking a color attribute value results in returning the product with the default thumbnail, which does not match all the possible items thumbnails. Or, searching for a searchable attribute color value results in retuning the products with a default thumbnail.

Overwrite the product's default properties, so that the thumbnail and other properties of the most relevant SKU match the search term. Therefore, the product is more visually relevant to the shopper's search term.


Scenario 5: Promote a representative SKU in search results

Business users can promote a particular SKU when multiple SKUs exist. For example, the business user could promote a SKU is on sale or overstocked. This scenario uses search rules and merchandising attributes.


Scenario 6: Promote a representative SKU in category pages

Business users can promote certain product-specific SKUs when a shopper navigates category pages, and before the shopper selects any facets. This scenario uses merchandising attributes.