Adding an effective search function to your e-commerce website isn't a routine "check it off" task — it's absolutely critical to your success. Online shoppers who use a search function are 1.8 times more likely to convert into a purchase, and 42 percent of those online shoppers begin their site visit by going to the search bar.
Today’s online merchandising is a blend of art and science focused on optimizing the e-commerce sales process. Recent advances in artificial intelligence (AI) are taking the guesswork out of the search function’s role in merchandising and allowing online retailers to use search in ways that directly drive sales.
AI’s Impact on E-Commerce Search
Website search functions traditionally use keyword-based technologies, which have issues dealing with imprecise search terms. For example, a shopper may search using "USB C" or "USBC" rather than "USB-C." To navigate the issues, companies have traditionally written hundreds of rules, created synonyms, or stuffed their sites with keywords as they try to account for every possible version or combination of key terms; it's a never-ending effort because the real world is constantly changing. Plus, rules start overlapping and conflicting with each other. It's so difficult to untangle the mess that has been created
Now e-commerce search can be implemented with AI-based solutions that don’t just look at keywords, but also understand context and user intent. They use that understanding to drive meaningful search results regardless of which keywords are typed in. The value to e-commerce retailers comes from better relevance and better ranking.
To illustrate relevance, imagine a searcher typing in "kettle bell" or "kettlebell" or "kettelbel" (typos happen more often than you think). Relevant results would offer exercise equipment, not teapots. An AI-based system knows the difference instantly.
The order search results appear matters, too. That’s called ranking, and AI can help here, too. For example, reinforcement learning is a type of AI that allows automatic optimization of rankings based on a range of buyer signals such as clicks, signups, conversions and ratings. AI takes the guesswork out of the ranking process.
How AI Informs Merchandising
The goal of online merchandising is to entice a browsing, prospective customer to "add to cart" and then complete the purchase. Traditional tactics include attractively designed catalog landing pages, manually ranking products in category pages by importance, adding sales banners to product thumbnails or pages, and product recommendations.
Speaking of recommendations, they can be one-to-one or one-to-many. One-to-one recommendations are personalized based on a visitor’s purchase history, browsing history, search history, or some combination of this information. Or, in the absence of a known visitor history, online merchandisers can use factors such as seasonality or product popularity to create one-to-many recommendations.
New AI software solutions are improving online merchandising’s effectiveness in influencing purchase behavior. The AI algorithms use collected sales outcomes to optimize product rankings and displays. Customer searches are tracked, along with resulting actions, like add-to-cart, sharing, purchase, and ratings. As more and more of these searches and actions are captured, AI uses the data to continuously improve results and put the highest converting products with the highest margins first.
The Intersection of AI Search and Merchandising
"Searchandising" describes the enhanced interaction between AI-based search and merchandising, an interaction directed towards helping e-commerce visitors quickly discover products that fit their requirements or match their needs. It works by dynamically curating both search results and dynamic product category pages for each online visitor based on their unique search, browsing and purchase history. Every visitor gets a personalized view of the website’s products.
AI-based search algorithms can also use rules based on any attribute — brand, rating, margin, date added — to automatically boost a vendor’s important products to the top of search results and category pages. This type of flexibility makes the search function an integrated component of marketing and merchandising campaigns.
Put Searchandising to Work on Your E-Commerce Site
You can harness the power of searchandising to create a thoroughly engaging search experience for your customers, an ongoing experience that nurtures brand loyalty, drives more conversions, and boosts revenue.
Joe Ayyoub is the chief revenue officer of Search.io, an AI-powered search and product discovery platform that helps e-commerce brands and retailers maximize their revenue and conversion rates through outcome focused, real-time search result optimization.
Related story: The Path to Purchase: It All Starts With AI Search
Joe brings over a decade of e-commerce and search experience to Search.io. Prior to Search.io, Joe served as chief customer officer at ZineOne. Before that, he was senior vice president of customer experience and partnerships at Unbxd and head of global support operations at Magento Commerce (acquired by Adobe). Joe holds a Bachelor's degree in Electrical Engineering from Jordan University of Science and Technology and an MBA in Finance and eBusiness from Golden Gate University.