How NLP Ecommerce Search Can Boost Your Revenue

How NLP Ecommerce Search Can Boost Your Revenue

Every year, almost 76% of consumers abandon a site after not finding what they’re looking for, costing ecommerce companies over $300 billion. But ecommerce retailers and grocers can retain this traffic (and revenue) by using natural language processing (NLP) search. We dive into the details below.

Table of Contents

How Natural Language Search Can Boost Revenue in Ecommerce 

Natural language search engines use artificial intelligence (AI) to understand the human language, efficiently and accurately returning products for difficult-to-understand queries. 

But let’s dive into the nitty gritty, outlining six ways NLP in ecommerce can drive your KPIs and boost your revenue:

1. Offering Autocorrect

Older ecommerce product discovery tools that aren’t built using NLP lack finesse with autocorrect. Their clunky computing power multiplied across millions of site visitors translates into lost revenue as misspelled queries produce zero results, and frustrated customers bounce.

A modern NLP ecommerce search engine draws on first-party data and AI to automatically learn language differences phonetically and typographically. It understands people’s common misspellings based on their distance on a keyboard and their varied pronunciations, even for unfamiliar terms.

It also provides features like Autosuggest (type ahead) and visual autocomplete—without manual involvement. This increases search success by helping customers find what they’re looking for faster, even if the exact keyword they type in isn’t featured in the product catalog.

Even an egregious misspelling of chocolate (“vhocolste”) on Target Australia’s site leads to chocolate products being returned along with visual autocomplete.

Misspelled search for chocolate on Target Australia fixed by Constructor product search and discovery platform.

2. Detecting Synonyms Better

Companies like Google and Amazon have successfully tackled synonym generation at scale. Others claim to bring you synonyms, but only fix spelling and punctuation—e.g., “ipod” and “I pod.”

Around 70% of search engines are unable to return relevant results for product synonyms, like “rain boots” and “wellies.” When customers leave after not finding the results they’re looking for, it’s known as shadow churn, and it accounts for approximately half of all new customers leaving a site.

To get beyond this issue, an NLP ecommerce solution processes the relationships between terms in a way that is not merely keyword-based. It uses robust recognition and substitution of synonyms to understand that “rain boots” and “wellies” are the same product.

And an NLP-based technology like Cognitive Embeddings Search (CES) learns from categories, product names, and text descriptions to solve the issue of frustrated and zero-result searches—even when very few search results pop up.

Zero-results showing up on Homebase's website for the search "iron skillet" despite iron products being available.

How does it work exactly? 

It combines data with deep learning to represent the product catalog as a “sea of stars” chart. Each catalog item is represented as a “star” on the same chart. From there, the algorithm measures the distance to each of its closest neighbors and makes an inference about what the user actually intended. 

The result is fewer frustrated searches and irrelevant suggestions, boosted revenue per visitor (RPV), and increased conversion rates (CRs)

Walking in Your Customers' Jimmy Choos

Speak their language. Improve your ecommerce search KPIs.

3. Reducing Search Difficulty

Not only does NLP-based technology reduce zero results and frustrated searches, it also solves a number of persistent issues that users commonly have with antiquated search engines, such as:

  • Product name uncertainty. Sometimes customers are unsure what to call a particular product. For instance, if they’re looking for a jacket with epaulets, they may type “jacket with shoulder flaps.” In such cases, an NLP-driven engine increases revenue by surfacing suggestions that customers are more likely to buy.

  • Long product names. When comparing prices between different retailers and grocers, customers copy and paste long product names. Older search engines return zero results because of their keyword-driven algorithms. But an NLP-driven engine removes unimportant words and acts on those that give material results.

    So if a customer finds “ALMO Men’s Regular Fit Organic Cotton Melange T-Shirts” on Amazon and copy-pastes that product name on your website, NLP returns a number of organic cotton t-shirts.
Target Australia website using Constructor Search and Discovery platform
  • Hyper-specific suggestions. Likewise, most search engines can’t associate generic searches with specific suggestions. If a customer in Minnesota types “not cold,” the engine should associate that with “warmth” or “winter,” surfacing specific suggestions that the user either accepts or uses as a jumping-off point for more refined searches.

    So a search like “I don’t want to be cold” on a site like Backcountry that uses an optimized internal search engine should provide something as specific as sherpa and polar fleece neck balaclavas in addition to insulated jackets.
nlp in ecommerce
  • Shortened spelling. A customer searching for “jmpsut” on a clothing retailer’s website doesn’t expect a zero-result page or irrelevant results. NLP search prevents them from abandoning the search by returning the intention of their query—jumpsuit.

4. Increasing Revenue from Long-Tail Queries

Long-tail queries are highly specific strings of more than three words entered by a user—i.e., “Castelli Warming Embro Cream.” They’re individually low-volume in nature (as depicted in the graph below), yet together they compose as many as 70% of internet searches. 

Because they’re often more product specific and reveal a higher level of purchase intent, they’re more likely to lead to a sale.

Search demand curve showing long tail search categories

While it doesn’t make sense to merchandise for these terms individually, using natural language search to return products for these queries is a huge untapped source of revenue. 

By intelligently mapping even unlikely search terms to products despite no keyword matches, NLP converts half-chances into real revenue at scale.

For instance, when a site visitor searches a query like “toys latest arrivals under $100” on Kmart Australia’s website, hundreds of relevant products are returned thanks to natural language search

nlp ecommerce

5. Offering Voice Recognition

The value of ecommerce transactions via voice search (think: Siri and Alexa) will grow to $19.4 billion by 2023. But this presents an interesting challenge for ecommerce retailers and grocers as voice queries are generally more colloquial than their written counterparts. 

For example, a text search query like “red lipstick” may be phrased as “Hey Siri, find me a red lipstick between 50 and 70 bucks” when voice searched. Because the product catalog isn’t reflective of this language, retailers may miss opportunities to return products for even the simplest of voice searches.

Luckily, an NLP ecommerce engine that automatically maps voice phrases, sorts product order by user intent, and applies filters can prepare you for a conversational search future.

6. Offering Multilingual Support

Nearly half ecommerce sites are multilingual, greeting customers in four or more languages. And some of the world’s largest companies—like Apple—generate more than 50% of their revenue from cross-border transactions.

While translation is important, communication is a two-way street. Does your search platform understand natural language search queries in every language you support? NLP is an ideal ecommerce solution in these use cases as it can “learn” any language.

Additional techniques like custom tokenization can specify how NLP should break each language down into discrete units. In most Western languages, we break language units down into words separated by spaces. But in Chinese, Japanese, and Korean languages, spaces don’t divide words or concepts. 

As such, custom tokenization helps identify and process the idiosyncrasies of each language so that the NLP can understand multilingual queries better.

English Japanese
Red Jimmy Choo shoes in size 8

The Future of Ecommerce is NLP

The ecommerce industry shows no signs of slowing down. Despite the pandemic, analysts project steady growth to $8.1 trillion worldwide by 2026. And with an NLP ecommerce solution powered by machine learning (ML) and AI, your company will be well prepared for the future. 

Give your users the pleasure of a great search experience using NLP in ecommerce as part of a natively holistic product discovery platform powered by AI and machine learning (ML). Not only does a forward-thinking platform help you successfully navigate tricky search results that would have otherwise led to site bounces, it also provides key insights that save your merchandising team precious time and helps you drive important KPIs to boost your bottom line.

No abandoned searches. No broken promises. Nothing but net.

This blog post was adapted from the ebook Walking in Your Customers’ Jimmy Choos: Optimizing Ecommerce KPIs with Natural Language Processing

Walking in Your Customers' Jimmy Choos

Speak their language. Improve your ecommerce search KPIs.