Optional filters - what they are and where to use them

Ecommerce sites have, for a long, used filters and facets to make search more efficient. Filters and facets allow users to narrow down search results based on specific criteria, such as price range, brand, or color. One of the key advantages of using filters and facets is that they can significantly reduce the number of irrelevant results and make it easier for users to find the products they are looking for. However, in some cases, traditional filters and facets may not be sufficient to provide the best search experience for users. This is where optional filters come in.

Optional filters are a type of filter that do not eliminate results based on certain criteria. Instead, they re-rank the query results based on whether they match or don't match the specific criteria. This allows users to see the most relevant results first, without missing out on other potentially useful results. For example, an ecommerce site may use an optional filter for “on sale” items, so that users can see the most discounted products first, without missing out on products that are not on sale.

A study by Econsultancy found that the use of optional filters can lead to an average of 15% increase in the click-through rate for search results. Additionally, a study by Nielsen Retail Group found that the use of optional filters can lead to a 37% increase in the average order value for ecommerce sites. These statistics demonstrate the potential for optional filters to significantly improve the search experience for users and increase conversions for ecommerce sites.

In this article, let's understand everything about optional filters, their types, how they differ from traditional filters, how they can be scored, the difference between simple and complex optional filters, and the caveats that need to be considered while using them.

What are optional filters? 

Optional filters are a type of filter that are used to provide more relevant search results to users by re-ranking them rather than eliminating them. They are designed to boost or penalize results based on specific criteria, allowing users to see the most relevant results first. This is different from traditional filters, which eliminate results that do not match the specified criteria.

For example, in an e-commerce site, a user might be looking for a dress in the color red and the price range of $50-$100. A traditional filter would eliminate all the dresses that do not match this criteria, making the user miss out on the similar dresses that are available in other colors or slightly higher prices. An optional filter could be used to promote the red dresses that fall within the price range, but also show the other options, like other colored dresses or the ones which are slightly more expensive.

Another example of an optional filter could be for a travel website, where a user wants to find a hotel that has a gym. A traditional filter would eliminate all the hotels that do not have a gym, but an optional filter can be used to boost the hotels that have a gym, but also show other hotels that do not have a gym but have similar amenities.

Broadly, there are two types of optional filters: positive and negative. Positive optional filters are used to boost results that match certain criteria, while negative optional filters are used to penalize results that match certain criteria. This way, the user is presented with the most relevant results while not missing out on the similar options.

For example, a positive optional filter for “new products” would promote results that are new, while a negative optional filter for “out of stock” would demote results that are out of stock.

Optional filters differ from traditional filters and facets in that they do not eliminate results based on certain criteria. Instead, they re-rank the query results based on whether they match or don't match the specific criteria. This allows users to see the most relevant results first, without missing out on other potentially useful results.

Understanding filter scoring

Filter scoring is the process of adding scores to different optional filters to further optimize the search results. Filter scoring works by assigning a score to each result based on how well it matches the criteria of each filter. The results with the highest scores are then displayed first, while the results with lower scores are displayed lower in the search results.

Filter scoring can be used to fine-tune the search results and make them more relevant to the user. For example, an ecommerce site may use filter scoring to promote results that are new, in stock, and on sale, while demoting results that are out of stock or not on sale. This can help to ensure that the most relevant results are displayed first, and that users are not presented with irrelevant or out-of-stock products.

Simple vs. complex optional filters 

There are two types of filter scoring methods: simple and complex. Simple filter scoring methods are based on a single criterion, such as the price of a product. Complex filter scoring methods, on the other hand, are based on multiple criteria, such as price, brand, and color.

Simple filter scoring methods are easy to implement, but they may not be as effective as complex filter scoring methods in providing the best search experience for users. Complex filter scoring methods, on the other hand, can provide more accurate and relevant results by taking into account multiple criteria. However, they can be more difficult to implement and may require more resources to maintain.

For example, a simple filter scoring method for a clothing ecommerce site may only take into account the price of a product, promoting products that are on sale and demoting those that are not. However, a complex filter scoring method may also take into account the brand, color, and size of a product, promoting products that are on sale, from a popular brand, and in a popular color or size.

Caveats while using optional filters

While optional filters can be an effective way to improve the search experience for users, there are some caveats to consider when using them. One of the main issues is the impact of optional filters on search performance. Optional filters can slow down search performance if there are too many of them or if they are too complex. Additionally, it's important to keep in mind that the more filters you have, the more complex the search query becomes, which in turn can slow down the performance. To avoid this, it's crucial to limit the number of filters used and keep them as simple as possible.

Another consideration is the use of standard replica vs. virtual replica. Standard replica is the default option, which includes all the data from the index, while virtual replica allows you to create a replica of the index, with a subset of data, that can be used for specific filters or queries. This can help to minimize the impact on search performance, but it can also be more difficult to maintain and update.

Additionally, it's important to note that optional filters can also impact the relevance of the search results. If not implemented correctly, they may demote or promote irrelevant results. Therefore, it's crucial to test and tweak the optional filters to ensure they are working correctly and providing the desired results.

In summary, the following pointers need to be kept in mind while using optional filters:

• Limit the number of filters used
• Keep the filters simple
• Be mindful of the impact on search performance
• Consider using virtual replica
• Test and tweak the filters to ensure they are working correctly
• Monitor the relevance of the search results.

It's important to weigh the benefits and drawbacks of using optional filters to ensure they are being used effectively and efficiently. To avoid these issues, it is important to minimize the number of optional filters used and to keep them as simple as possible. This can help to ensure that search performance is not affected and that the search results are still relevant for users.

Conclusion

In conclusion, optional filters are a powerful tool for ecommerce sites to improve the search experience for users. They allow users to see the most relevant results first, without missing out on other potentially useful results. However, it is important to consider the caveats when using optional filters, such as their impact on search performance and the use of standard replica vs. virtual replica.

One such brand that makes the use of optional filters easy is Zevi, an AI-powered filters and faceted search solution. It uses advanced algorithms to optimize the search results, providing users with the most relevant results in the shortest amount of time. Zevi offers several benefits for both consumers and brands. For consumers, it allows for natural language search, making it easy to find products regardless of the quality of their description in the catalog. Brands can also benefit from Zevi by rearranging their sort order based on their business goals and customizing the user experience throughout the customer journey. Additionally, brands can use customized discount labels to attract customers and also address mixed query languages like Hinglish and Devanagari. If you're looking to improve your ecommerce search experience, consider Zevi as a solution. To learn more, book your demo today.