Relevance

In this sprint, you will configure relevance within Algolia, using features in the Algolia dashboard.

Relevance webinar
Register for Webinars and view the Recording

Team members

Depending on the size of your company, some of these roles may be the same person. This sprint it is important we identify these roles and get in contact with them.

Project Manager

Planning and project oversight


Product Manager

Product vision, planning, prioritizing and management lifecycle

Merchandisers

Management of e-commerce product display

Searchable attributes

By default, when you upload your data to Algolia, all attributes are searchable. This is unlikely to be the optimal set up, as some of  attributes are only to be used for display, ranking, or faceting, and some are more important than others. Follow this guide and this documentation to understand which attributes to make searchable and how to order them


Custom Ranking

During the Data Design sprint, you identified and indexed business data so that search could be more relevant. You can configure business relevance through the custom ranking setting, which breaks the ranking for records with the same textual relevance.


Custom_Ranking

Setting language

If you indexed data so that there is  one index per language, you can optimize relevance by setting removeStopWords, ignorePlurals (this can be done directly or by setting the index) index language, and query language.

For Danish, German, Finnish, Dutch, Norwegian, Swedish, you can set decompoundedAttributes and decompoundQuery. If decompounding is not working as expected,  you can customize it through custom dictionaries.

Algolia supports language transcriptions through the customNormalization setting.

For example, in German, umlaut vowels (“ä”, “ö”, “ü”) are commonly transcribed with “ae”, “oe”, and”ue” if the umlauts are not available on the keyboard. In the same way, “ß” can be transcribed as “ss”.

setting language

Iteration and review

Relevance should never be viewed as ‘finished’. There are always optimizations that can be made. It is recommended to frequently test top queries, and if they return less desirable results, to troubleshoot them.

Rules

Rules let you make precise, predetermined changes to your search results. You should use them for r edge cases and optimizations. There are many use cases where Rules are relevant, but they shouldn’t be the core strategy of your relevance configuration.

Synonyms

Synonyms tell the engine which words and expressions to consider equal. They come in many different configurations.

Personalization

Personalization could be a good fit for you if your users tend to return and want to access similar items, for example books from the same author, clothes from the same brand or documents from the same project.

To implement personalization you need to ensure that you are sending a unique userToken with queries. This can be checked in the "Implementation help" tab of the Personalization page in the Algolia Dashboard. If all events are valid, and the userToken with queries is present, then you can enable personalization and set your  personalization strategy. It is recommended to A/B test your personalization strategy before going live for all users to see how effective it is.

Personalization_Sport

AI Synonyms

Algolia identifies queries that your users often change and proposes synonyms for them. You can accept and reject based on your understanding of what your users are searching for.

AI synonyms

Dynamic Re-Ranking

Dynamic Re-Ranking is an Algolia feature that leverages AI to find trends in your users’ behavior. Based on the query and the position of the result they click or convert on, it can make improvements to your relevance by boosting results that are rising in popularity. To enable Dynamic  Re-Ranking, you need to validate your events. Then you can test your reranked queries in the Re-Ranking Simulator. Once you are happy with the results, you can launch an A/B test and then go live for all users.

AI re-ranking