What is the best way to evaluate your search and discovery experiences?

Great search and product discovery is a competitive advantage. 

Customers return to sites that offer rich experiences. How do you know if your site search is best-in-class, or if your competitors outperform you? 

The quality of site search revolves around the basics you probably already know (like relevance and speed) - but more importantly, it revolves around goals you may have heard less about, like the attractiveness of results, personalizing those results in the right places, and being precise about what business objectives product discovery should optimize. This is in addition to other more subtle metrics we will also discuss. 

So, when evaluating your search and product discovery against competitors, where do you start?

Table Stakes Functionality

These facets of product discovery optimization are table stakes. Optimizing them won’t ensure your product discovery will compete with more advanced retailers; however, they do lay the groundwork for great customer experiences. 

1. Speed.

How do your search, browse, recommendations, and autocomplete speeds compare to your competitors? 

Research at Amazon showed that every 100ms of added load time cost them 1% in sales. Google found an extra .5 seconds in search page generation time dropped traffic by 20%. Conversion rates and customer experiences both improve as latency decreases. 

2. Platforms.

Where and how do users access your product discovery experience? Do your competitors serve consistent experiences across more devices and media? 

Customers want a coherent experience - the look and feel of your site should be familiar on every device. Customers, who are increasingly shopping on mobile devices, should not feel they’ve entered a different store when they move from a laptop to a phone. 

3. Basic relevance and attractiveness.

How relevant are search results on your site compared to your competitors’ sites? Do you know if customers are finding what they are searching for on your site? 

While relevance will ensure ‘shoes’ don’t appear when users search for ‘laptops,’ it leaves out an important part of the equation: attractiveness. Attractive search results are results that have the highest chance of leading to an increase in a business metric (conversions, AOV, etc.). We’ll discuss this more later in the article.

Advanced functionality

1. Search bar features.

Does your site search use typo-tolerance, plurals, and natural language processing (NLP) to help your users find what they’re looking for? Do competitors offer more search bar features to help users find products quickly? 

An effective search bar does more than predict basic user queries. E-commerce giants can combine NLP algorithms with autocomplete to present query results that most closely approximate user intent. These algorithms are commonly used to:

Correct phonetic misspellings, keyboard proximity typos, punctuation nuances, and omitted character typos. Is it “blue ray”, “blue-ray” or “bluray?” The correct spelling is “blu-ray”, but you shouldn’t require users to know this to get good results.

• Detect synonyms to serve relevant results to users. If a user is searching for “suntan lotion,” good autocomplete will recommend “sunscreen.”
• Rank word importance to better understand what results will most likely lead to a conversion.
• Add support for multiple languages to give the best search experience to any visitor.

2. Navigation.

Poor navigation can run customers away just as easily as ineffective search. When users can’t find the products they’re looking for, they assume they aren’t sold on the site. They do not persevere and many never return. 

Faceted navigation pays major dividends. Once you understand what product features your users care about most, you can display them on the page so users don’t have to work to find them. Facets should fit the context of the search query (e.g. if a user is searching for “blazer”, you may decide to rank facets like “fit” and “jacket style” higher in the navigation). If they’re searching for “pants,” facets like “inseam” and “front style” would be ranked higher.

3. Personalization.

Do your users get personalized search results based on their preferences as demonstrated by their behavior on your site? Do your competitors personalize results to users? 

Modern personalization as a search feature can revolutionize customer experience. You may think that personalization means showing different results based on standard demographics or some other gross descriptive category; however, great personalization allows you to improve the experience on your site for every individual down to minute details (ex. using affinities towards organic foods for grocery retailers, or golf pants for apparel retailers). 

Where is personalization being deployed on your site - is it being used across the board, or only in some parts of the user experience? 

The revolutionary aspect of modern personalization is that it can be fine-tuned to increase business key performance indicators (KPIs) as well as to improve user experiences. Note that “over-personalization” can actually deliver less relevant results. Check your competitor’s site - If you search for a product and click on a result, and then begin seeing that type of product everywhere (ex. searching and clicking on a shoe, then searching for shirts and seeing shoes appear for the shirt query), be wary. Authentic personalized search learns from customer behavior and returns uniquely customized results.

Staying competitive in the future of product discovery

While it may be difficult to determine which exact functionalities are deployed on competitor sites, you can assume many will be moving to add them. If you seize this opportunity to get an early competitive advantage, you will future-proof your business, and will reap the benefits of higher conversions and improved customer loyalty.

1.Does clickstream data fit into your ranking strategy?

We define clickstream as data on the actions each user takes on your website — what action each user takes — including searching, browsing a category page, clicking, adding to cart and purchasing. We’ve developed a unique process that provides retailers tangible insights into their discovery gaps and opportunities.

Clickstream data forms the foundation for great search and discovery algorithms like Learn to Rank (which applies machine learning to relevance) and personalization. Clickstream is integral in ranking products; Great search systems automatically re-rank search results for products most likely to lead to a conversion. An example of results re-ranking is as follows:

If 40% of users who search for “laptops” purchase the laptop in the 8th search position and 22% purchase the laptop in the 5th search position, it would make sense to re-rank the 8th laptop higher (perhaps in one of the first few positions) and the 5th laptop closely following. Other laptops that users don’t frequently purchase should be moved down in the search results.

Beyond its benefit to users’ search objectives, clickstream ensures your users are served attractive search results -  results optimized for valuable KPIs like conversions.

2. Business KPIs.

Are your search and product discovery experiences optimized for business KPIs?If they are not, you should identify business metrics that you want site search to drive.

Leading search companies like Google and Amazon optimize their search results to be attractive to each user and tie it to a particular business KPI to meet its objective. This KPI can be any number of things, e.g. conversions, margins, click to conversions, add-to-carts, etc. By optimizing your search for a particular KPI, you are able to generate search results that benefit both you and your users.

3. Machine learning.

Is machine learning being leveraged in your search and discovery experience?

In the past, most search and discovery systems required hours of manual work to ensure results were (at a minimum) relevant. But search science has moved beyond simple relevance based on user keywords and human tweaking of ranked results. Companies like Google and Amazon optimize their search results to meet selected business KPIs, and so can you.

Machine learning, beyond delivering better results to the customer, ensures all your search results are optimized to drive important business KPIs. It also allows you to implement advanced search systems like personalization, that will provide users richer experiences and give your business a significant competitive advantage.

4. Connected product discovery experiences.

How much do your product discovery channels, e.g. recommendations, browse, and search, learn and share data with each other? 

You or your competitors may have all the functioning pieces of a complete product discovery experience, but are they connected? Are users having a consistent experience across all the facets of product discovery? Are the signals your users provide to you being used across the board? 

An investment in search and product discovery can transform customer experience. We have demonstrated that optimizing product search only for “relevance” will almost certainly not result in products that both delight your users and meet your KPIs. Modern search and discovery can do so much more.

Benchmark your competitors’ sites for the features we have described. You will find ways to distinguish your product and your site. Beyond delivering better search results for your customers based on their behavior across your site across platforms, you can optimize results based on key performance indicators that are sure to improve the bottom line.