When e-Commerce site search works well, it is fast, convenient, highly relevant and positively impacts the bottom line. Many sites don’t pay enough attention to search, noting that less than 10% of site visitors typically utilize it. In general, the impression is it wouldn’t be wise to spend much time optimizing for such a small subset of users — but search is different.
Site visitors using search are your hottest leads, often in the final stages of the buying process, having contemplated and made their decisions, and now just looking to execute their purchase. While the average e-Commerce conversion rate is just 1.72% for non-search visits, visitors that interact with search at least once have a conversion rate 3.5 times higher, at 5.98%. On average, visitors who use search to navigate an e-Commerce site will convert at a rate 348% higher than those who do not interact with search.
That’s why we view search as the most under-optimized site feature capable of yielding the highest returns. Here are five key tips to maximizing the power of search:
Advertisement
Natural Language Processing (NLP) Focused On Product Awareness
Search technologies like Google Site Search, SOLR and ElasticSearch were originally developed to mine large databases of files. These were based on keyword matching, displaying all documents containing one or more words in a single search term. This type of search was good for content search, but not the wide variety of product searches consumers use today. Additionally, these technologies lack intelligence, which is left up to developers to code and maintain.
Today’s online shoppers simply do not have much patience, and on-site search must strike the critical balance between delivering the most relevant results and not overwhelming searchers. The key to this balance is NLP that is geared more towards the specific way e-Commerce site visitors shop (with short terms and phrases) — understanding the primary item and purchase context and discarding (or deprioritizing) less relevant or non-relevant results. This is known as product awareness.
Take the example of a shopper looking for “black flared dress.” NLP understands that “dress” is the primary item being searched for, and if something is not a dress, that is a deal-breaker — no matter how many matching qualifiers there may be (for example, black flared pants). NLP also learns from previous searches which types of words are the most important qualifiers (descriptors) for various products. For example, for dresses it may be style, while color is secondary. So looking at “black flared dress,” the search engine will recognize “flared” as the style and prioritize a blue flared dress over a black mini dress in the rankings.
Personalized Site Search
According to Accenture, 75% of consumers are more likely to buy from a retailer that recognizes them by name, recommends options based on past purchases, OR knows their purchase history. This makes personalization a high priority for e-Commerce sites.
Intelligent site search is a prime vehicle for personalization, “learning” from individual shoppers’ earlier inputs to tailor search results in accordance with unique shopper tastes — for example, a preferred color or style. A shopper may search for a particular jacket, and based on previous searches, the engine “knows” which color the shopper likes best and will display the jacket in that color. This approach makes e-Commerce sites much more effective at appealing to shoppers and inciting them to purchase.
Automating Synonyms, Misspellings And Redirects
When conducting searches, nothing annoys site visitors more than receiving a “no results” page. The goal must be minimizing these pages and this rests in large part on search engines’ being able to recognize and learn common synonyms and misspellings. For example, site search must be able to produce boots, even if the search term used is galoshes.
This may sound obvious, but it is surprising how many e-Commerce sites lack this capability. According to a Baymard Institute study, up to 70% of the 50 top-grossing U.S. e-Commerce web sites require visitors to search by exact jargons. For instance, if “all-in-one printer” was typed, these sites would not be able to also return relevant products labeled as “multifunction printer.”
Search must also be able to address misspellings, which are common and can ultimately hurt a site’s ability to close sales if not managed. According to a recent eBay analysis, words like sheer (“shear”), elegant (“elegent”) and striped (stripped) are among the most frequently misspelled words in site searches. To address common misspellings, e-Commerce sites can create landing pages with a curated set of products and set up redirects. So for instance, any search for “stripped shirt” will still return a selection of striped shirts.
Optimize The Search UX
It’s impossible to completely avoid “no results” pages, and there are steps that can be taken to enhance the search user interface, increasing the likelihood that shoppers will “stay in the store.” One example is confirming the search — confirming and reminding shoppers of what they searched for, as well as providing recommendations like checking for spelling mistakes, or trying different or fewer keywords.
Other techniques for optimizing the search UX include persistent search queries (once a search is conducted, the keywords remain in the search box on the next page). By saving the original search query in the text entry field, site visitors can more easily modify their search (over 80% of these modifications are simply the addition or removal of a single word), increasing the chance they’ll find what they’re looking for.
Additionally, sites can provide sample text in search boxes, as well as search scope selectors — once shoppers conduct a search, a pop-up window prompts them to specify certain attributes or categories, like brand, color, size or gender. Ultimately these UX optimizations help shoppers search more successfully.
Analyze And Adjust In Real Time
Every day, site visitors are providing a wealth of data that they can put to work to enhance competitive edge, and a lot of this information comes from search. What types of products are customers most interested in? What search terms are generating zero results, such that product findability can be improved?
Additionally, which products on your site are under exposed, over exposed, top performers or trending the most heavily? This type of information can help optimally determine marketing spend allocations. This data can help identify areas for investment and guide site design decisions, such as which features should be prioritized and displayed most prominently. A highly profitable e-Commerce business is always a work in progress, and the dynamics driving conversions at any point in time are very fluid. Organizations must be able to sense and appropriately respond to these changes in as close to real time as possible, and analyzing search data is just one key to this.
Conclusion
Because site search users have the biggest impact on your profitability, catering to them is important. The tips described here are a good first step, and in the future, we expect search to play an important role in other profit-boosting initiatives, like online merchandising and personalization. An unoptimized site search with poor relevancy confuses visitors, while an intelligent, highly relevant site search brings shoppers over the conversion finish line.
Trevor Legwinski is Chief Strategy Officer for SearchSpring. He is a creative and strategic thinker with more than 10 years in the retail industry, working for Cambria Cove, Strands, and in-store operations for Abercrombie & Fitch. In this time he has also managed e-Commerce and marketing for Hallmark, CycleGear and Bambeco. Legwinski has deep expertise in what it takes for retailers to succeed in online commerce as well as brand, product and offline marketing. He holds a Bachelors degree in Marketing from St. Norbert College and an MBA in Marketing from Oregon State University.