E-commerce

Why discovery is critical for today’s e-commerce search engines
facebooklinkedintwittermail

Few e-commerce search engines actually do their job: helping customers easily find what they’re looking for. The key word here is “easily”. The frustration involved in the overall product search experience results in an unacceptable level of churn and burn: to the tune of 68%, according to Forrester.

The same exact percentage—68—applies to the number of websites found to offer poor search result sorting options, and a full 72% of sites completely fail site search expectations.

The expression, “You had one job!” comes to mind.

Clearly, we need better e-commerce search engines. But the remedy isn’t necessarily to make e-commerce search engines more like, well, search engines. It’s about making them more like—that word we’ve seen popping up everywhere since around 2015—experiences.

And that’s where discovery comes in.

Search vs. search and discovery: what’s the difference?

E-commerce search happens when the customers know exactly they’re looking for—e.g., a “four-foot-high black bookcase” — and the e-commerce search engine finds it, or at least attempts to find it.

This relationship worked fine when e-commerce was just getting started. There weren’t many items to choose from on any given e-commerce site, and sites were overall offering mostly transactional experiences: either your desired product was there or it wasn’t. But as the volume of data and the sheer number of products available online grew exponentially, e-commerce search grew increasingly difficult for companies to get right, and increasingly frustrating for users to navigate.

Take the above search term, “four-foot-high black bookcase”, as an example. Enter those specific words into the top left home page search bar of one of the world’s biggest e-commerce furniture retailers, Crate & Barrel, and you get two somewhat contradictory messages. One is that there are no search results at all, the other that there are in fact 440 results for the search term “bookcase”. 

No search results on an e-commerce search engine

Alas, those results are a mish mosh of items similar to a four-foot-high black bookcase but not really that at all. The first row, where you’d expect to see your item, features a few very tall brown bookshelves and a short white kids’ bookshelf.

wrong results for e-commerce site search

 

Furthermore, when you click on the “type” filter at the top left, what comes up isn’t what you would hope and expect: the option to filter by height. Instead, the very first option is, confusingly, “bookcase”, and the other options are items barely related to bookcases, such as entertainment centers.

At this point, the customer looking for the four-foot-high black bookcase will most likely bounce for a quicker, more intuitive e-commerce search experience.

Instead of an overwhelming labyrinth of semi-relevant, semi-related items that the searcher isn’t interested in and doesn’t need, a great search and discovery experience would not only provide precise results for the exact search term, but also pull up a streamlined interface of highly contextual recommendations and filters, personalized to the particular searcher’s age, geographic location, device being used, previous visits to the site, and much more. This would make the searcher feel known and understood, and make it easy for them to find not just the black bookshelf but other things they previously may have not realized they needed.

The tools of e-commerce search and discovery

How does search and discovery accomplish this?

Via a set of tools and techniques that work together to provide fully contextual, enjoyable e-commerce search experiences that make searching adventurous and educational instead of laborious and transactional.

illustration of e-commerce search and discovery tools

Let’s dive in.

Leveraging the search bar

The search bar itself can be a window to discovery. You’ve most likely encountered an experience where you type in a query, and a dropdown box appears with suggestions of more specific or more popular queries.

autocomplete pattern in e-commerce search

This autocomplete experience — also known as predictive search, type-ahead, autosuggest, or search-as-you-type — gives users search suggestions as they are typing into the search bar, using context to predict what they might be looking for. Autocomplete can boost sales and conversions by as much as 24 percent.

Navigation

Great e-commerce search and discovery includes powerful and intuitive browsing experiences that can easily work without the search bar. Think about Netflix, where the entire experience is discovery-based.

In the e-commerce realm, this could be something like the experience on Artsper, an arts buying and selling website that will offer browsing pages via clicking on one of the top navigation categories, such as “Painting”. The “Discover” category provides a literal discovery experience that can lead users to unexpected and interesting buys they may have not considered before.

discovery experience on Artsper's e-commerce search engine

Filtering and faceting

Most of us are familiar with search filters, which enable us to narrow down search results to categories applicable to what we are looking for. For example, we could narrow down a search for a headset by choosing a brand or a category.

filters on an e-commerce search engine

 

But, when we search for a bluetooth headset specifically, our experience is that much better if filters offered are narrowed down to the ones applicable to bluetooth headsets only.

example of facets on e-commerce search engine

 

This is what facets do: they make sure that only filters that match the result set are available to you. In addition to making for a smoother, faster discovery experience, facets prevent the dreaded “no results” page, which could happen if the same filters showed up for your “headset” and “bluetooth headset” searches.

Filters and facets are often used interchangeably; for a great explanation of nuanced differences between the two, check out this article by the Nielsen-Norman group.

Providing a unified search and discovery experience

You should allow your users to be both searchers or browsers.

Searchers are in pursuit of finding exactly what they are looking for (e.g., a pair of jeans in a certain brand, color and size). Browsers, on the other hand, have a potential intent to get something, but may not exactly be sure what (e.g., a business casual outfit that may include a pair of jeans).

Your e-commerce site search should enable users to start by either searching or navigating, then further refine their results by leveraging the other strategy in addition — for example offering facets after the user performs a search, or making filters and facets searchable.

Here is an example of using a category page search box to enhance the filtering/faceting experience.

searching category pages in e-commerce
https://www.algolia.com/doc/guides/solutions/

Business ranking and merchandising

In addition to providing your users results most relevant results based on purely textual matching, your e-commerce business has a great opportunity to offer them content or products they are most likely to act on. For example, you could track conversion rates on products, and display first results with the highest conversion rate, or customize rankings based off your product popularity, newness, or availability. With this ability to modify search results based on your business metrics — what we call “business ranking” —  you can also test which products your customers want, and refine rankings further based on the experiments’ outcome.

This is where merchandising comes in. Use your search and discovery tool to reflect a promotional campaign, merchandise keywords like “t-shirt”, merchandise category pages, pin certain items, or hide items from results.

hiding items in search results

Recommendations

Recommendations can give searchers a set of matching results, brands and categories that allow them to easily drill down into what they want. Ideally, these recommendations are incorporating key user data, such as their past search history or demographics. This is also an opportunity for e-commerce businesses to offer sponsored results, items with high margin, etc.

Amazon does a great job here (although one could argue that their recommendation algorithms could do better at times):

Amazon's recommended items
https://www.algolia.com/doc/guides/solutions/gallery/related-items/

Geo search

Geo search, also known as location-based search or local search, is another powerful e-commerce search and discovery tool that improves the search experience by letting customers know which products of yours are close by and available.

The power of e-commerce search and discovery

Using search and discovery in a unified, fluid way will give your customers a clear, intuitive search experience that will shorten their path to the shopping cart, and leave them satisfied with your site — and happy to return for more.

Consider this: e-commerce is now happening everywhere. Social media platforms like Instagram and Facebook have native shopping tools, and personal assistants allow consumers to make purchases using their voice. The more powerful and easier it is for users on your digital properties to find both what they’re looking for and what they may want—even though they may not yet know they want it—will make your e-commerce search engine the business-driving juggernaut it is meant to be.

About the authorIvana Ivanovic

Ivana Ivanovic

Senior Content Strategist

Recommended Articles

Powered by Algolia AI Recommendations

The (almost) ultimate guide to site search
Product

The (almost) ultimate guide to site search

Ivana Ivanovic

Ivana Ivanovic

Senior Content Strategist
How search and discovery helps you stay ahead of e-commerce trends
E-commerce

How search and discovery helps you stay ahead of e-commerce trends

Matthieu Blandineau

Matthieu Blandineau

Sr. Product Marketing Manager
How to choose the right e-commerce site search tool
E-commerce

How to choose the right e-commerce site search tool

Matthieu Blandineau

Matthieu Blandineau

Sr. Product Marketing Manager