What goes into a great search experience?

The Elements of Search [Infographic]

15 Best Practices for 2022 to Improve E-commerce Site Search

15 Best Practices for 2022 to Improve E-commerce Site Search

The pandemic has put the e-commerce industry on steroids with an ever-increasing number of e-retailers vying to reach consumers who are ready to by online. One way to stand out is to improve customer experience through better search usability and product discovery on your site. This article will cover 15 different ways that you can do just that!

  1. Make the search box more visible
  2. Run autocomplete and autosuggest
  3. Improve search conversion rate optimization with machine learning
  4. Better relevance and accuracy
  5. Better ranking
  6. Search personalization
  7. Use synonyms
  8. Use search analytics and data to improve experience
  9. A/B testing
  10. Manage misspellings
  11. Add filters and search facets
  12. Leverage natural language processing (NLP)
  13. Make it mobile-friendly
  14. Offer voice search
  15. Don’t forget product recommendations
  16. Bonus: Pick a great e-commerce site search solution

Make the search box more visible

Your website visitors should be able to find your search box easily. Here's why: studies show that visitors are 216% more likely to convert than people who only browse your site. 

Many e-commerce sites only use a small search magnifying glass icon. To assist and encourage visitors to search, it's a good practice to design a search bar that's visible from any page on your site by placing it at the top of each page with an icon that stands out and clearly communicates what you can do when you click it.

Run autocomplete and autosuggest

Autocomplete, also called autosuggest, predicts what searchers are looking for as they type to provide smart suggestions. One study found that buyers were 6x more likely to convert because of autocomplete functionality.

Autocomplete has many benefits that might not be obvious at first:

  • Offers a list of search suggestions to help buyers narrow their results faster
  • Provides visibility into products that your buyers may have not seen otherwise
  • Can correct for misspellings to ensure visitors are more likely to find what they're looking for
  • Gives you, the seller, an opportunity to merchandise results

It's no wonder e-commerce giants like Amazon, Walmart, eBay and Best Buy have all added autocomplete suggestions on their site. And you can do the same quite easily with modern site search solutions... more on this below.

Improve search conversion rate optimization with machine learning

Your site search results page design, product images, titles, and descriptions, and other search UI elements can have a big impact on search revenue.

However, search success begins with good search results. At the risk of stating the all-too-obvious, if searchers get poor results, they're going to bounce. Conversion rate optimization begins with delivering more relevant results and improving result ranking.

Here are a few ways you can optimize results:

Better relevance and accuracy

Search relevancy is a measure of how close the search result is to the search query. Inaccurate results reflect poorly on your brand and drive customers away from your site. Most visitors do a poor job of reforming their original queries. You only get one shot, which is why machine learning-based accuracy is important.

Getting accurate, good results can be tricky. If someone searches for “queen bed fitted sheet,” it means they’re looking for sheets, not mattresses, but search engines don’t always understand that. 

To improve results, you can add rules to improve accuracy. Traditionally this was accomplished by writing hundreds of rules — and synonyms, keywords, and product tags — to make it easier for search engines to find content.

With new machine learning algorithms, that effort can be cut back by 95% or more. You can deliver more relevant results even when site visitors use synonyms, write long queries, or even describe what they're looking for. This is precisely why we developed Neuralsearch — to handle the thousands of normal search queries and edge cases to deliver great results. 

When evaluating search providers, be sure to inquire about AI relevance models to assist with search accuracy.

Jakob Nielsen quote

Better ranking

Ranking and relevance are related and can influence one another. Whereas relevance is the accuracy of a result, ranking is the order of results. Top results get more engagement than results at the bottom of the list — and exponentially more clicks than results buried on page two or three. 

A search for “fitted sheets” could return 10 equally relevant results, but how they’re ranked is important. To boost conversions and deploy merchandising campaigns successfully, you’ll need to control ranking. 

New machine learning solutions such as reinforcement learning can boost results automatically based on the criteria you specify. With Search.io, you can also dynamically boost results on attributes such as rolling 30-day best selling, highest ranking, or highest margin, or some other combination of rules to drive great results.

Get Personal with search personalization

Consumers want and expect their favorite shopping sites to be personalized. Search personalization was once something only the Amazon's of the world could manage, but affordable solutions are now available to any merchant.

Personalized search includes search results that are tailored to each individual based on their profile, which can include past search history, purchase history, brand preference, product ratings, gender, and more. Personalization solutions can also improve results based on browser IP information such as location; even if a visitor is  new or logged out, results can be delivered one-to-one.

Site search solutions like Search.io can be used for e-commerce personalization and recommendations: 

  • Search results can be ordered or optimized with personalized products
  • Product catalog pages can be personalized
  • Recommendations can be included throughout the buyer’s journey

Use synonyms

You say tomato, I say solanum lycopersicum. Well, not really, but synonyms are part of your customers’ vernacular. If someone searches for "bag" on your site, they might be looking for tote bags, suitcases, backpacks, or something else. If you have a solution like Neuralsearch that understands the meaning and context of search queries, synonyms are automatically created for you. However, if you have an older generation of search solutions, they should offer a way to manually create and manage synonyms.

search analytics

Use search analytics and data to improve experience

It’s fascinating how many articles about improving customer loyalty include “surprise” experiences — such as flash sales or better emails. In fact, the “boring” experience of finding exactly what you want is a better start! 

Search metrics can help you improve on-site experience in several ways:

  • Use the search keywords your customers are using for product titles and descriptions. Your data has a host of valuable information. The way people shop is how you should sell.
  • Identify search trends to uncover shifting buyer trends, identify highest and lowest conversion rates to improve your site taxonomy, merchandising, and product descriptions.
  • Review and attend to poorly performing search results asap before they impact sales.

A/B testing

You can not only A/B test results page layout and design, but also search results. Test different search configurations to find a winner. This is how Catch.com.au boosted search revenue by millions of dollars per year. Using Google Analytics or search metrics from your search provider, you should be able to get a good sense of how well your site search is performing.

Manage misspellings

Misspellings are one of the most common e-commerce search issues. Autocomplete is important in early search interactions, and unexpected suggestions from minor typos may prompt buyers to look for different ways to browse, modify queries, or abandon your site due.

Despite this being a well-known issue, it's still a big problem for companies. Baymard's benchmark report shows that 69% of major sites still don’t support autocomplete spelling suggestions for even slightly misspelled search terms.

Bottom line: Find a search solution that manages misspellings.

Add filters and search facets

Filters and search facets are a great way to get visitors down the funnel towards the specific product they're searching for. For example, common filters include product category, color, pricing, and size. 

Some best practices for filters include:

  • Multi-filter selection: Allow visitors to select more than one filter e.g., size, color and style. If your site and search are fast enough, results can be update din real time. However, if it's slow you'll want to give people a chance to batch filters before refreshing results.
  • Filter hierarchy: Put the most important filters above the fold, e.g., you might make brand most important then price. This will help prioritize their purchase decision-making process.
  • Breadcrumbs: Help your site visitors to adjust search results by moving by changing their filter settings. In other words, use breadcrumbs to show how they arrived at the search results e.g., size, color, style, etc, and a way to easily adjust filters.
  • Search within filters: If filters are too long, you can allow site visitors to search within them e.g., you might offer 40 brands of shoes and rather than forcing users to skim the list, you could give them a way to search and select the brands they want within the filter.

Leverage natural language processing (NLP)

E-commerce website search can be improved by leveraging natural language processing (NLP) to help users find what they're looking for e.g., if a visitor types in "black, dress shoes women's size 11," it may not return the results that were expected and lead them to leave your e-commerce store.

In that example, a search engine needs to understand each part of the query — is the buyer looking for a "black dress" or "black shoes that are dressy"?

NLP algorithms can help e-commerce search engines understand what a shopper is looking for. They can identify the intent behind a user's input and return results that are more relevant to their query.

Make it mobile-friendly

Half of all internet traffic shops from a mobile device. So what does this mean for e-commerce sites? Well, it's time to make your e-commerce site search responsive and optimize it so that users can find products from a phone or tablet with ease.

voice search

Offer voice search

Voice search is growing in popularity e.g., Amazon has reported they've sold "tens of millions" of Alexa devices, making it one of the fastest-selling device ever on their site.

Users are increasingly relying on voice assistants to help them shop e.g., asking Google Assistant or Siri for recommendations and reviews before deciding what to buy. Many retailers are already investing in voice search in how product pages are designed. The same optimizations you're doing for internet voice search can be used for on-site search.

Don’t forget product recommendations

Your search provider can not only power relevant product results, but product recommendations as well. Surface related products or "People also bought" items on relevant results pages to drive additional revenue. 

The best e-commerce site search solutions not only help you optimize site visitors' intent, but are able to merchandise your catalog as well so that it's easy for buyers to discover the right products.

Bonus: Pick a great e-commerce site search solution 

Bonus recommendation: pick a great search provider like Search.io to transform mediocre site search into a search machine. We built our platform from the ground up on entirely new search engine technology to offer extraordinary machine learning, fast response times, and easy-to-configure settings. Features such as autocomplete suggestions, spell checking, filter creation, and more are available out of the box.

Regardless of which provider you choose, however, just be sure to invest in search. Better search leads to higher version rates, a better customer experience, and repeat visitors. 

Signup for a free 14-day evaluation or schedule a time to speak with one of our search experts to learn more.

Similar articles

User Experience (UX)
Best Practice

Drastically Improve Your Website's UX by Improving Site Search

Best Practice

An Introduction to Custom Search Engines

Best Practice

How to Get a Competitive Edge with Smart (and Surprisingly Easy) E-commerce Site Search Solutions