Reference information - Search Preferences Types

Last updated:
Edit on GitHub

There is a set of search preferences’ types that you can specify for your attribute key. All of those types possess different features. In this article, you can find out the detailed information about the search preferences.

Full Text

Full text implies that, if set to Yes, the attributes will be included for full-text search. It means that a user will be able to find products when they search for a text which is present in the value of a searchable attribute.

Example The focus_adjustment attribute key has the following values:

  • Auto
  • Auto/Manual

If Include for full text is set to Yes for this attribute, then, when typing any of the values of this attribute in the Search field of the online store, all words will be searched, and the full phrase will have a higher weight than separate words from it. Meaning, if Auto/Manual is typed, the results having Auto/Manual value for the Focus Adjustment attribute will be displayed first in the search results flyout: Full text{height=“500” width=””}

Full text boosted

Full text boosted implies that, if set to Yes, the attributes will be included for full text boosted. It means that the attribute values of these specific attributes will receive a higher relevance than other attributes having the same values.

Example The alarm_clock and waterproof attributes both have Yes and No attribute values, but:

  • alarm_clock has the Include for full text search value set to Yes
  • waterproof has Include for full text boosted set to Yes

In this case, when typing Yes in the Search field on the web-shop, the products that have the Waterproof attributes with the “yes” value will appear higher in the list of results than products with Alarm clock attributes with the same value. Full text boosted{height=“500” width=””}

Info
If several attributes having the same values have been included for full text boosted, they all will appear in the search results. Their order is provided by Elasticsearch and can be further improved by customizing its analyzers.
*** ## Suggestion terms

Suggestion terms implies that, if set to Yes, the attributes will be included for a suggestion. This search preferences type implements the “did you mean” search functionality which provides alternative suggestions when a user may have misspelled a search term.

Example The storage_media attribute has the SSD and Flash values. If Include for suggestion has been set to Yes, then when a user types flashs in the Search field, the search results page will contain a box with suggested search term “flash”. Include for suggestion{height=“200” width=“500”}

And vise versa, if you don’t include an attribute for the suggestion, when a user searches by its values and misspells them, there will be no result for the user’s search. Do not include for suggestion{height=“200” width=“500”}

Completion terms

Completion terms implies that, if set to Yes, the attributes will be included for completion. It means that typing a word in the search field brings up predictions, which makes it easy to finish entering the search form.

Example The storage_media attribute has SSD and Flash values. If Include for completion has been set to Yes, then when user types “fla” in the search field, the search term will be autocompleted with “sh,” and there will be a list of suggested terms in the search results flyout: Completion terms{height=“200” width=“500”}

Tips and tricks When you first decide on activating search preference types for attributes, keep in mind, that enabling all of them is highly unrecommended, as this will result in a huge list of search results. Instead of this, you might want to consider enabling search preference types for only those attributes that you really want your users to find while searching, or the attributes that refer to products you want to appear in the search results above all.

Example
There is a new device in your shop which is popular on the market for its video recording properties. You know that users are very interested in a device with such property and they might search for products by it.
Suppose, you have created the _video_recording_ attribute in your shop with the values **Geotaging** and **Autofocus**.
However, suppose there are other attributes having the same values.
Since you want to advertise the specific new device more, it would make sense for you to disable, or at least to restrict the number of active search preference types for all other attributes with **Geotagging** and **Autofocus** values and enable an individual (or even all
search preference types for the _video_recording_.
This way you will make the _video_recording_ product attribute searchable and therefore the products with this attribute will stand out in the search results when your customers search by attributes.)

Also, it does not make much sense to activate search preferences for attributes with th enumeric and Yes/No values. As numbers may occur not only in attributes but in product SKUs, names and descriptions (which are actually ranked higher than attributes in search results), therefore the probability that a user will find what they were looking for is low, but the list of search results will be huge, and the search term will be present in multiple places. Besides, it is very unlikely that users will be searching for an attribute with a numeric value or the Yes/No values.