Episerver is committed to help you come back stronger. Learn more about our COVID-19 response.

All customer success stories Content Cloud customers Commerce Cloud customers Financial services B2B manufacturing B2B distribution B2C retail Commerce Cloud Content Cloud Intelligence Cloud Digital Experience Platform Content Intelligence Web Content Recommendations Web Product Recommendations Customer Data Platform Cloud services Headless All capabilities Financial services B2B manufacturing B2B distribution B2C retail Resource library Customer Hub Trust Center Content management Ecommerce Personalization Email marketing CCPA Digital Experience videos Articles Guides Reports & Studies Events Webinars Customer success stories Partner Network App Marketplace Training & Certification Support Expert Services Managed Services Customer Hub User Guide Tips and tricks Developer documentation Quick start for developers Developer community

Charles Tyrwhitt

Machine learning significantly increases sales

Bringing in-store recommendations to online shoppers

A majority of the business for British retailer of men's shirts and gentlemen's apparel, Charles Tyrwhitt, is generated online.

The company decided to enhance its website to better reflect its overall “customer first” policy. A key aim was to recreate the personalized shopping experience enjoyed by customers in-store for those buying online.

In a shop, personalization is delivered through dialogue between staff and the customer to find the perfect shirt, followed by suggestions on a tie or cufflinks to complement it. If a product is out of stock, the sales assistant understands what the customer wants and can recommend the ideal alternative. This means customers get value-added services and Charles Tyrwhitt optimizes its sales opportunities.

Online, recommendation solutions can take on this role, but many fail to generate real understanding, making uninformed customer comparisons and assumptions. Charles Tyrwhitt needed a low-maintenance solution that delivered real insight into what customers want, encouraging cross-sales even when their first choice of product is out of stock.

Personalization based on each visitor’s behavior

Charles Tyrwhitt looked at a range of recommendation providers, then selected Episerver Product Recommendations. The solution was trialed on the website over a period of three months, with results showing that it would deliver an estimated additional £1 million in annual sales.

“Basing recommendations on what other customers have done is not good enough, because we know that no two people are the same,” says Jennie Blythe, Head of Web Development and Trading at Charles Tyrwhitt. “Episerver’s technology understood this, with individual insights replacing standard-issue algorithms.”

Episerver Product Recommendations uses more than 70 algorithms, as well as hints and filters to analyze a customer’s entire engagement history as they explore the website. While the technology is complex, implementation is easy, injecting product information into the website using a line of javascript. The solution works as soon as it’s plugged in, and it also enables the company to customize the best algorithm for a particular website placement.

Episerver holds quarterly review meetings with Charles Tyrwhitt to ensure that they are delivering maximum value. “We knew that out-of-stock items were impacting sales performance, so we wanted the technology to come up with a solution to tackle this,” Blythe says. “The team took it in their stride, customizing the product recommendation engine so that it supplied appropriate alternative product recommendations that would both satisfy customers and secure what would otherwise have been a lost sale.”

Personalized alternatives to products that are out of stock

Today, when customers come to the Charles Tyrwhitt website, they not only receive product recommendations that complement their desired purchase but intelligent suggestions if their initial choice is unavailable.

Episerver delivers quantifiable, bottom-line benefits: the solution increased average order value by 36 percent in six months; and it increased average units per order by 24 percent in six months. Optimal recommendation positioning of the basket delivered 28 percent of recommendation revenue, and 6 percent of recommendation revenue came from the out-of-stock feature.

“In essence, product recommendations bring our customers a personalized shopping assistant, mimicking the in-store retail experience,” says Blythe. “It means that customers enjoy a more satisfying online experience and get the products they really want, while we optimize sales and build a trusted relationship with individual consumers. It’s a recommendation service we can recommend.”

In essence, product recommendations bring our customers a personalized shopping assistant, mimicking the in-store retail experience.

Jennie Blythe

Head of Web Development and Trading Charles Tyrwhitt

Project Facts
Charles Tyrwhitt uses Web Product Recommendations

Episerver uses cookies and browser technologies for analytics, personalized content and ads.