Today’s standard retail search and discovery experiences are falling further and further behind shopper expectations, primarily being driven by non-retail technology advancements.
Take, for example, the next-generation semantic search capabilities used by Google, Apple and Wal-Mart, the highly engaging visual shopping and personal content curation provided by Pinterest, as well as the enlightening and smart music recommendations provided via relevant back-and-forth communications between Spotify and its users.
These are the new standards for engaging with shoppers and retailers are faced with some steep challenges in evolving to meet them.
Shoppers and merchandisers do not share the same vocabulary
Shoppers have been trained to think in short keywords, guessing at a translation between their preferences and needs and how they think the manufacturer and merchandiser may have categorised a product. Or, they must navigate via a highly structured, yet limited set of criteria, often times flicking between product pages and the navigation bar digging for information to help them find the right product for them.
At the heart of this limitation is a lack of “humanised” merchandising data or terminology.
For example, a shopper might describe a skirt as “fuchsia, pencil” and the retailer might call it “pink and slim fit.” The shopper might also want this skirt to be appropriate for an interview.
Very few retailers have the resource to look too far beyond manufacturer-provided information to determine the myriad of ways that their shoppers truly think about each aspect of a product.
The result is a sub-par set of product merchandising meta-data that can’t be used by a user to efficiently and confidently discover the products best for them.
Merchandising is not continually optimised
Most merchandising typically involves a “set it once and forget it” process.
Retailers are strained to understand what information has the most impact on buyer decisions, as well as what types of information and criteria are completely missing from the shopping discovery experience.
Early detection of changing trends, styles and shopper vocabularies must be facilitated to help the merchandising team recalibrate and prioritise optimisations.
Natural language discovery and navigation are insufficiently supported
Current search and discovery tools have been built to only interpret structured keywords rather than shopper “thoughts” and concepts.
Even if more rich merchandising data was available, current tools would not be able to effectively capture shopper needs and preferences in a natural, semantic way and translate them into highly relevant results.
As with data gathering, the resource required to update taxonomies and filter options fueling product discovery is time-consuming and often heavily dependent on development support and roadmap prioritisation.
Personalised merchandising requires more than implicit data
Great strides have been made over the past ten years to create algorithms intended to help anticipate and point shoppers to products and content they are likely to find interesting and useful.
As important as this implicit data can be for retailers and brands, the gold standard is explicit data, such as preferences and needs directly expressed from shoppers, as it can play a more significant role in learning from each shopper’s likes and dislikes and buying path to improve the overall shopping experience.
As TechCrunch reported, “While retailers are doing more with the implicit data (i.e. reminding you when you left items in your shopping cart, etc.) no-one has yet mastered the art of capturing that precious explicit data.”
The key to the next wave of online shopper-driven discovery is not in adapting the experience to the user, but also letting the user adapt the experience to his or her preferences. It is important for retailers to move beyond the idea of static experiences or static content, and instead embrace a philosophy that gives shoppers the tools they need to make decisions faster and more confidently.