The Retail E-Commerce Guide to Product Discovery

We’ve all been there. One of the easiest ways to get people within the retail and e-commerce industries animated, waving their hands and talking loudly is to poll them on their own personal experiences trying – and failing – to find what they’re looking for when shopping online.

There’s the search for the “fitted holiday party dress” on the specialty fashion website that turned up “men’s dress shirts” instead – and zero dresses, fitted or otherwise. 

There’s the major luxury retailer’s recommendations engine that decided to recommend teacups and hand towels to the shopper who wanted something to pair with the collared checked work shirt in his basket. 

Why is this? Why does product discovery so often go haywire? What’s clear is the impact of poor product discovery – both on the consumer and on the retailer who’s not able to connect them with the products they’re looking for. The 2022 Google/Harris Poll tells us that a whopping 94% of consumers say they’ve dropped their online shopping session because of irrelevant search results. These consumers are all too willing to leap off of one retailer’s site and onto another in hopes of finding those shorts, that dress, that pillow and that eye makeup that matches their personal style and taste.

In this guide we’ll take a look at where product discovery flounders, and attempt to illustrate how it is product attributes – those descriptors and multiple features that make a unique product unique – that, implemented correctly, bridge the gap between a merely mediocre product discovery experience and a world-class, high-conversion, customer loyalty-driving experience. 

You’ll walk away from this brief foray into product discovery with a strong understanding of just how much better it can be when it’s truly infused with the language of the customer – the way they/we actually describe and search for the products we care about.

Your Shoppers Control the Narrative

Word of mouth travels a lot further and a lot faster than it once did. Frustrated shoppers who can’t easily find what they want have no qualms about telling others about it on social media. They keep retailers’ social media and PR teams on their toes, no question – but how much of the root problem is actually being addressed when these shoppers vent their concerns online?

This sort of unwanted publicity can be, if not fatal, a true narrative-shifter in how consumers look at a retailer’s brand. Not only are these consumers frustrated with their shopping experience, they’re pulling others along with them – and it doesn’t have to be that way. Let’s take a look at where it all starts.

The traditional path to attribute newly-arrived products is the one that retailers have long relied upon: hire a few associate buyers or an outsourced agency, and ask them to do this manually. These buyers will often take the legacy, out-of-the-box attributes that come directly from the manufacturer and distributor, and use those as the basis for their own item set-up. 

What’s wrong with that, you ask? Well, nothing, as long as all of your customers are searching exactly for the most vanilla attributes imaginable: colors, sizes, brand names and merchandise type. A “red men’s large t-shirt” should be no problem. A “red men’s crew neck long torso linen blend logoed shirt”? Go ahead and try it on your favorite retailer’s website, or something else that more fits your own personal style. This is where the legacy, manual attribution model often breaks down, and breaks down badly.

You see, the problems that start at item set-up have ramifications all the way across product discovery:

• SEO and SEM: They can’t find your products on Google, neither via paid or organic search, because they’re not searching for generic, legacy attributes – so they’re not arriving on your site in the first place
• On-Site Search: Even once they arrive, you’re not matching customers to the products they’re actually looking for in site search

Product Recommendations: Retailers are therefore unable to personalize & provide recommendations that convert or further increase order sizes without understanding the relationships between attributes

This is because manual, legacy attribution results in thin and inconsistent product attributes. Customer-centric product attributes, on the other hand, allows for trends, styles, subjective and personalizable attribution that helps to capture the “long tail” of shopper searches, both on SEO/SEM and in your existing site search engine.

Manual attribution, often performed by short-tenured employees or outsourced business process organizations with little-to-no skin in the game, is very often the weak link in the chain – with large ramifications where it counts.

Every consumer has their own personal way to describe what it is they’re looking for, and they bring this to bear when they’re looking to be connected to the products they want. As we make our way in the world, we learn different colloquialisms, expressions and features that we use as shorthand to describe what’s important to us. 

Yet this “search diversity” is barely acknowledged on a high percentage of modern e-commerce retail sites, and the result is missing results from on-site search that, had they been returned to the curious shopper, could have led to an instant purchase. 

Remember: different shoppers search uniquely. You’ll want to adapt and expand your product taxonomy to capture as many of them as you can. Take a look at the synonyms for “bracelet” in this example. What you might call a band I might call a armlet; your best friend might call a bangle and her mother might call it a wristlet.

The key to serving all four of us is to build a robust product taxonomy that takes into account how real people search in the real world. Sure, those who are already focused on a particular brand name can often unknowingly filter their search just to the results from that particular brand (i.e. Nike, Hugo Boss, Topman etc.), and that certainly makes it easier. For the majority of shoppers looking to compare items between brands, however, it’s essential to assign attributes to items that allow for the rich and varied diversity in what we actually call things, and what’s truly important to us when we shop.

Retail e-commerce brands capitalize on shoppers’ search diversity when they start enabling products to be surfaced in a language of customers, not merely legacy, out-of-the-box attributes. Your customers know what they want, you just have to show it to them!

So now we know that relevant product data powers relevant search results. Yet this isn’t just on a retailer’s own website – it’s way up in the chain, right when a customer pops open their favorite search engine to look for something they want to potentially buy. What if you could drive dramatic increases in relevance and conversion “up the funnel” with search providers like Google and Bing, and ensure that more high-purchase-intent customers land on your site, and not someone else’s?

Traditionally, retailers who rely on manual attribution provide thin and inconsistent product details to their PDP pages. This keeps customers from finding those products on Google and other search engines, and has the marketing team bidding on ineffective keywords that don’t drive traffic to the site.

Search engines and product listing ads check sites and pages for how well the content on them match a user’s search terms; that user’s shopping intent, and a whole host of other factors. When you use the same keywords in your content that your target audience is searching for, this signals to search engines that you’re giving the people what they want. Keep this relevant and up to date, and search engines will reward you for getting it right. 

It’s simple: the better product attribute data you can proactively provide to a search engine for your SEO marketing and paid e-commerce listings, the more consumers will find that data in their searches – then buy what you’re selling. Adding customer-centric product tags directly to your PDP page (title & description) is a start. Including structured attribution directly in your Google Merchant feed gets you even further, as does applying relevant keywords to backlinks on your e-commerce site. 

Furthermore, retailers and brands who work with Lily AI receive the benefit of scalable, automated keyword recommendations and analytics. Retailer-identified keywords can be run against Lily AI product data to see relevant attributes and the number of products within the catalog aligned to those keywords; retailers can adjust accordingly, right in the Lily AI platform.

On-site search is perhaps the most powerful product discovery use case. They’ve intentionally made it to your site – now it’s your opportunity to accurately connect them with the relevant products they’re looking to buy. 

Up to 43% of visitors to retail websites immediately go to the search bar, and those consumers are 2-3x more likely to make a purchase than site visitors who don’t use the search bar. And yet nearly 84% of companies don’t actively optimize or even measure their on-site search, which points to a true opportunity in the marketplace. It’s no longer acceptable to return 10-plus pages of results that drop off dramatically in relevance after the first click. The retailer who does invest some time and money into optimizing their site search can stand out quite quickly.

This is again where enhanced product attributes – infusing your existing site search engine with the language of the customer – truly pays dividends. This is measured in boosted click-through rate and increased search-to-PDP conversion. Get them to the product detail page, you’ve won most of the battle. They’re interested! 

Recommendations are another important component of retail e-commerce product discovery, and one that most retailers invest in heavily. That said, they also want to make sure that shoppers are being presented with items that truly match the attributes that make up their personal style and preferences, providing them all the right reasons to make that second click and to boost revenue per visitor.

Delivering better product recommendations starts with understanding how product attributes intersect and interact with each other. For instance, an expanded taxonomy that tags multiple products as “Boho chic” or “coastal grandmother” or “party” – rather than just black, pants and medium – will match shoppers to the products they actually want. Recommendation relevance increases when you present a product set that’s been built by understanding how shoppers really describe the products they’re looking for – as well as the relationships between product attributes.

With the language of the customer now a part of your product taxonomy, you can now increase the number of relevant products shown to each customer – and just as importantly, stop recommending items that don’t match what they truly want (as in our embarrassing teacups and hand towels example earlier in this guide).

In 2020, a luxury retailer that works with Lily AI began a deep look into some of its key e-commerce business metrics, and had a hypothesis that conversion from online searches was lower than both the company’s own expectations and below industry norms. In setting out to understand why, the company sought to understand precisely what the company’s shoppers were looking for. The company found that while many searches were focused on either distinct brand names or on broad categories (i.e. “dresses” or “shirts”), over 50% of searches were attribute-heavy, long-tail searches that combined materials, trends, occasions and other customer-centric attributes.

For instance, they found that their shoppers were often searching for “sequin” in the retailer’s on-site search engine, and were returned between 10 to 20 sequined dresses; and yet the company knew that they had over 200 sequined dresses that could have been returned. When the company dug into why this was happening, they came to realize that its manual, human-based, merchant-driven attribution was a clear bottleneck in terms of consistency, speed and accuracy, and in ultimately allowing for better conversion from search queries on its e-commerce properties. The company wanted to reach these high-intent customers, and knew that Lily AI could help.

The company’s SVP of Customer and Revenue Growth had this to say in a 2021 webinar:

“Then we saw something rather remarkable. There were two metrics that were really interesting. The first one was, of the many people that I show these better attributed results to, how many people click into a product page? That increase was kind of mid-single digits. Super interesting to me, because again, these are like really high-intent customers, who are now finding a lot more of interest for them to click into. Now I’ve kept them on the website longer, and I am getting them to a product page; so further down my funnel. 

Even more interesting was the metric of: are they actually converting on those products, and they were converting even more. I saw that if I could show them things that were more relevant, they’re going to click on that; they’re going to spend more time on my site exploring, and they’re much more likely to convert. So this was hugely interesting, because it told me, basically, there’s a lot of lift for my overall business for any product where I could attribute them in this more reliable way. I’m engaging customers by showing them more relevant products, and I’m helping them complete their shopping journey, which is good for them and good for me and good for their value to me in the long term.” 

By using Lily AI’s expanded product taxonomy and language of the customer to augment their on-site search, the retailer has increased the relevant results for descriptive searches by up to 30x. This translates to $20 million in incremental revenue for the company, which led to a further expansion of the Lily AI product attributes platform across the company’s diverse, varied and rapidly-growing product catalog.

Better Product Discovery with Lily AI  

By now, you’ve certainly gained a sense of where the product discovery process breaks down; the importance of customer-centric product attribution; and how this bolsters a retailer or brand’s existing efforts in SEO/SEM, Search and Product Recommendations.

Let Lily AI show you how to deliver the most extensive, accurate and consistent product attributes that match the language your customers actually use. No other platform allows marketers, merchandisers, product owners and digital teams to instantly boost their on-site conversion, enable rapid product onboarding, sell more inventory at full margins and drive 8-9 figure revenue lift like Lily AI’s product attributes platform – all in one centralized place. Contact us now for a 30-minute demo, and we’ll show you how it works!