Salesforce Expands ‘Smart Search’ Program For Retailers

"We’re moving customers from the search box to checkout quicker using best practices for keyword hypernyms and synonyms,” said Grace Edinger Manager, Digital Product at Kiehl’s.

A year after Salesforce launched its artificial intelligence platform Einstein within its larger CRM Commerce Cloud system, the company is promising retailers “smarter search” in the latest update.

The rollout of Einstein Search for Commerce is being  embedded across the entire Salesforce Platform, powered by machine learning, deep learning, predictive analytics and natural language processing.

Einstein Search for Commerce makes search smarter in several ways, says Gordon Evans, VP Product Marketing:

“Einstein Search Dictionaries consumes all site searches and surface terms that are being searched for but are not in retailer’s keyword list,” he says. “It then recommends a synonym to add to the list. Previously, if a member of the e-commerce team wanted to identify missing search terms and choose which existing synonyms list to add, they would be forced to sift through a lengthy spreadsheet and then guess which synonym list would be most appropriate.”

Further, Einstein Search Dictionaries analyzes data across the Commerce Cloud platform to identify relationships between search terms, making recommendations on which synonym list admins should assign them to, Evans adds.

GeoMarketing: Is Einstein Search for Commerce a new capability or an advancement of previous capabilities?

Gordon Evans: The features of Einstein Search for Commerce are new to the Salesforce Commerce Cloud, and will make search smarter in several important ways.

First off, Einstein Search Dictionaries recommends search terms for retailers to add to their synonym list by surfacing terms that are being searched for but are not in retailer’s keyword list for their website.

For example, a shopper searching for “mauve sweater” might not get any results, but Einstein Search Dictionaries would recommend that the brand add “mauve” to the synonym list for “pink” and “purple.” The benefits for shoppers and retailers are clear. Shoppers aren’t disappointed, coming away empty-handed when they can’t find the product they want.

Retailers benefit by eliminating the time-consuming and inefficient process of manually going through search logs, finding terms and then guessing where they should be added, as well as not losing a sale to a customer who can’t find something they actually have in stock.

Another key feature being announced is Einstein Search Recommendations, which personalizes typeahead (or search-as-you-type) search, suggesting search results for both known and anonymous shoppers. With each click and interaction, the engine gets smarter. It uses the shopper’s behavior, such past and recent searches to power 1-to-1 recommendations, helping consumers find the products they want (or those they didn’t even know they wanted) much faster. Before a shopper even types a single letter in the search bar, they are presented with drop down search recommendations, just for them.

Finally, Commerce Cloud is also launching Keyword Search Sorting Rules to give merchants more granular control over how product search results appear on the site, empowering them to override ‘standard’ results.

Merchandisers can prioritize certain terms and pull specific results forward to highlight products, and create custom assignments so specific search terms return results driven by sorting rules outside the default.

For example, if a shopper searches for “sandals” the results are sorted by new arrivals, or a search for “skincare” might be sorted by ‘top rated.’ Previously, regardless of whether a shopper searched for sandals or skincare, they’d only see results driven by the default sorting rule.

We talk a lot about “micro-moments” — searches that people on-the-go make to satisfy an immediate need (“best coffee near me,” “clothing shops open now,” for example) — will these new features help Salesforce clients better capitalize on those kinds of searches?

Einstein Search for Commerce is focused on helping shoppers find the products they are looking for on a brand website more quickly and gives retailers smarter search tools for their websites.

For example, Kiehl’s is leveraging the Commerce Cloud and search to drive sales, and has already seen shoppers complete the checkout process more quickly.

“We’re moving customers from the search box to checkout quicker using best practices for keyword hypernyms and synonyms,” said Grace Edinger Manager, Digital Product at Kiehl’s.

Will these features work differently across search engines? Or is it mainly focused on Google? Or does it have nothing to do with Google?

These new features are not relevant to Google, but rather to a retailer’s website. For retailers, search is a gateway to their products, driving up conversion rates and revenue.

According to our Commerce Cloud Shopping Index quarterly analysis of more than 500 million unique monthly shoppers, consumers who use site search are 2.4 times more likely to buy. Searchers also drive more revenue, spending 2.6x more across mobile and desktop retailer sites compared to those who don’t use search. In fact, the 9 percent of shoppers who use search across Commerce Cloud customers, drive nearly a quarter of their overall sales.

As shoppers today spend less time per visit on brand sites, especially on mobile, giving retailers smarter search tools can have a major impact on sales.

About The Author
David Kaplan David Kaplan @davidakaplan

A New York City-based journalist for over 20 years, David Kaplan is managing editor of A former editor and reporter at AdExchanger, paidContent, Adweek and MediaPost.