Configuring per-category strategy

Define per-category merchandising strategy to create and enhance brand experience

Why should you do it?

Applying manual merchandising with rules, such as pinning, hiding, burying or boosting products or categories allows you to override the standard strategy and have the flexibility to customize the brand experience for a specific use-case and make fast changes based on your business needs. These include, promoting a specific brand, pinning to the top an item that is currently trending on social media. Additionally, when implementing facet merchandising and ordering strategies you are able to customize the product discovery experience on each page and leverage business data to improve conversion rates.

Manual Merchandising with Rules

Manual Merchandising is a useful tool you can leverage to apply very specific changes and strategies on top of your global elevance and automated merchandising tools. 

Do: use manual merchandising as the last step of your relevance strategy, for one-off promotions, specific campaigns.

Don’t: avoid substituting global relevance strategy with large amount of manual relevance rules. This will lead to multiple bugs and conflict and limit your scalability and use of advanced tools and strategies.

For example:

When your relevance strategy is set to rank results based on the amount of user likes - more popular products will be displayed higher on the category page. Your new socks collection for Valentine’s Day is launching tomorrow and doesn’t have any likes yet. Therefore, you would want to override this ranking strategy and pin specific new products to the top of the product list on your category page or boost the whole Valentine’s Day collection. 

When a merchandiser's intent is to create a brand experience, which will reflect the brand’s values or positioning, you may choose manual curation strategy. For instance, if your company wants to emphasize products supporting a social cause that is aligned with the brand’s values, you may want to manually place these products in specific positions on the category page, instead of relying on an automated strategy promoting products based on their popularity, margins , and conversion rates on alternative ranking strategies. You can create rules, such as pinning products in chosen positions on a specific category page.

For a step-by-step instruction on applying manual merchandising strategies on Search results and Category pages, please refer to Business User Guide to Relevance and Merchandising in Algolia Documentation.

To better understand the concepts and the logic behind Search results and Category page merchandising strategies, please refer to the corresponding Algolia Academy section on Concepts of Search and Category Merchandising.

Facet ordering and Merchandising

Facets allow users to narrow down search results in order to discover items of interest. In several cases, you would want to make changes to the types of facets presented on specific pages, facet value or order to improve relevance and conversions. 

  • Dynamic facets: On large e-commerce websites/marketplaces, the variety and number of facets is too large to display all of them all the time. Another important aspect is that for products from different categories, many facets will not make sense. 

For example, users will not expect to see the same filters for books vs. computer screens when they refine their search. 

  • Ordering facets and facets value: You will need your facets or their values to show in a specific order in your search page.

Example: choose the “brand” facet to appear first, followed by product time, price and customer rating. Hide the “sale” facet, when the promotional sale period is over.

  • Facet value ordering: for each facet use the dashboard to perform the following functions
    • Pin one or multiple values at the top of the list
    • For the remaining values, specify whether they should be sorted alphabetically, by count, or not be displayed at all.

Example: The facet values for the facet “size” are ordered by count, which makes very little sense for the company's customers. The solution is to define in the facet configuration a custom order so that the values are always returned in the order “XS, S, M, L, XL”

  • Facet merchandising: tweak and adapt the facets ordering and the facets values ordering for a given query, category, or context in the Visual Editor. This functionality allows to override the settings within specific context or for a specific category.

Example: Your business data shows that the brand “Apple” tends to convert better than other brands. For my facet “brand”, pin the facet value “Apple” at the top of the list to increase conversion rates.

For a step-by-step instruction on applying manual merchandising strategies on Search results and Category pages, please refer to Business User Guide to Relevance and Merchandising in Algolia Documentation.

Takeaway 1


Takeaway 2


Going further

Additional Resources

Inspiration Library

UI practices, suggestions, etc. Discover more real companies' tactics to reduce no result rates