Filters

Filters allow the refinement of search results, simplifying the product exploration process and helping customers to find the right products faster and more efficiently.

Filter types

Ecommerce Search has two filter types to narrow down search results:

• Term filters for filtering textual data.
• Range filters for filtering numerical data.

Term filter

Term filters are used to include or exclude items based on specific terms. For example, if a user is looking for a specific brand or product type, term filters can be applied to ensure that only items that match those criteria are displayed.

A term filter consists of a key specifying the field to filter, a list of terms, and a match type specifying how the filter should operate. If the match type is not specified, Any is used by default.

1
2
3
4
5
6
7
8

{
    "filters": {
        "color": {
            "values": [ "blue", "white" ],
            "match": "any"
        }
    }
}

1
2
3
4
5
6
7
8

{
    "filters": {
        "weather": {
            "values": [ "waterproof", "wind-resistant" ],
            "match": "all"
        }
    }
}

Range filter

Range filters are ideal for numeric data where a user may be interested in items within a certain range. This could be price, size, weight, or any other quantifiable attribute. Range filters can be set with minimum and maximum values to define the bounds of the search.

A range filter consists of a key specifying the field to filter, from and to values, and a match type specifying how the filter should operate. Either from or to can be omitted, for an unbounded upper/lower range.

1
2
3
4
5
6
7
8

{
    "filters": {
        "size": {
            "from": 41,
            "to": 43
        }
    }
}

Handling missing fields

Missing field behavior specifies how Ecommerce Search should handle documents that are missing the filter fields during searches. This is useful when using Unified search, which returns both products and content that do not share all fields. For example, content does not have a stock or price field, but products do.

When defining a filter you can specify the missing field behavior. The default behavior is ignoring the document.

Applying filters

Ecommerce Search has two main mechanisms for applying filters, facets and domain filters.

Filters can be represented by the following diagram:

Facets

Facets are user-facing and provide a way for shoppers to interact with the search interface to refine their results.

Facets are essentially attributes that can be used to dynamically narrow search results. They are typically displayed on the search results page and allow users to filter results based on categories such as color, size, brand, etc.

Ecommerce Search has two types of facets, which mirror the two available filter types:

• A term facet lists all possible values for a given field. It also indicates whether the facet is selected or not, and which specific values from the facet are selected.

• A range facet contains min and max specifying the value range of the products. It also shows the selected from and to.

Facets are applied by adding a filter that matches the configured context facet to the search query.

Facets must be created on a context before they can be used. Creating facets consists of the following steps:

1. Ingest products with attributes
2. Add the field you want to use for the facet as an allowed filter
3. Add the field as a context facet on a given context
4. Do a search that resolves to that context

1
2
3
4
5

{
"segmentId": "b2c-dk-da",
"scopeId": "Browse",
"phrase": ""
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35

{
    "facets": [
        {
            "key": "COLOR_ATTRIBUTE_ID",
            "name": "Color",
            "facetType": "Term",
            "values": [
            {
                "count": 4,
                "isSelected": false,
                "key": "BLUE_ID",
                "name": "Blue",
            },
            {
                "count": 4,
                "isSelected": false,
                "key": "GREY_ID",
                "name": "Grey",
            },
            {
                "count": 4,
                "isSelected": false,
                "key": "ORANGE_ID",
                "name": "Orange",
            }
            ]
        },
        {
            "key": "Price",
            "facetType": "Range",
            "max": 1300,
            "min": 500
        }
    ]
}

Domain filters

Domain filters are used to apply broad filtering rules that may not change often. For example only displaying products that are currently in stock or available in a particular region. Domain filters are usually predefined by the store's search configuration

Unlike facets, domain filters are not directly visible to the user, but are applied behind the scenes, determined by scope, to ensure that search results are relevant to the store's domain-specific rules.

Domain filters cover the following filter types:

• Domain filters specified on a scope
• Filters applied to Authenticated search
• Filters applied by the client

Filtering on predefined fields

Ecommerce Search has some predefined fields on content and products/SKUs. The following table provides an overview of filterable predefined fields. Note that not all of these have filtering enabled by default, see Allowed filters for additional information. It is also important to note that the field names are case sensitive.

1
2
3
4
5
6
7
8

{
    "filters": {
        "CategoryNames": {
            "values": [ "Makeup" ],
            "match": "all"
        }
    }
}