Shoe-d-vision Personalization

Delivering a highly personalized commerce experience

In a strategic partnership with Shoe-d-vision and Raptor, IMPACT A/S built a new digital platform to serve as the heart for Skoringen’s future commerce experience.

The result is a best in class digital commerce platform with Umbraco Cloud as the central application. Since the new platform was launched, there has been an increase in mobile conversions of 23%.

At the same time, drive to store increased by 49%, meaning that a higher percentage of store visitors are now qualified to make a purchase.

This performance showcases a platform that fulfills Shoe-d-vision’s ambitious plans for creating a personalized user experience across several channels.

Behavioral data is tracked and activated on communication channels through several Raptor-personalization services, including web recommendations and behavioral e-mail triggers.

Finally, all user data is tied together with the Raptor Customer Data Platform, which unifies customer data and makes it accessible to other systems. The CDP centralizes customer data from all sources and enables Shoe-d-vision to act on all of their combined behavioral data.

Visitors that land on product pages will find relevant alternatives right below the product description. Shoe-d-vision display recommendations based on similar and recently viewed footwear on product pages.

If the initial product the customer landed on did not fit their needs, the recommendations help to inspire and convert browsers to buyers. This is illustrated below:

Below you see an example of similar products recommended as alternatives to the product above.

These recommendations are based on a combination of variables, including the user’s previous behavior and overall browsing patterns.

In the above case, the browsing pattern made up of clicks on other products from similar categories is given the most weight in the output. For a product to classify as similar, the clicks pattern has to occur within the same session.

This gives the algorithm an indication that the current product has a relation to the recommended products.

Adding the “Similar products” module to a product page will increase the conversion rate and decrease the bounce rate.

On almost every section of the Skoringen webshop, users can always find their recently viewed products. This provides the user with easy and personalized navigation to products that might still be of interest.

 On skoringen.dk, users that engage with the recommendations (Raptor users) have a 26% higher conversion rate than those who do not. Raptor users also generate an average revenue that is 21% higher than non-Raptor users.

Basket abandoners are encouraged to convert with automated email triggers

Shoppers that abandon their basket are often interrupted in their purchase or not entirely ready to make a decision. Personalized abandoned basket emails, containing exact products that were abandoned, represent a great way to increase conversions and recover lost sales.

These emails can be enriched with personalized product recommendations, that expose your customers to products that are often purchased with the basket content, products that are on sale, or a combination of multiple variables.

Raptor Customer Data platform (CDP) ties customer data together

The CDP provides a single customer view and allows Shoe-d-vision to build actionable audiences based on data from all customer touchpoints.

The audiences are activated in their existing tech stack. An example is their email platform, and Raptor’s personalization services on their owned and paid media channels.

Re-engaged past buyers through data-driven and dynamic audiences

The audiences range from simple to more advanced in their logic.

The audience below identifies shoppers that have previously visited skoringen.dk in a given period, without making a purchase.

Skoringen can also use The Raptor Customer Data Platform to identify its most customers, that have a high Customer Lifetime Value.

The audience can be activated in the email platform through the Raptor-integration, which pairs the individual subscribers with their previous and current behavioral data.

Using that information, we can recommend products based on previously visited or purchased products. This can be combined with products that are often bought with the customer’s last visited or purchased product, and that makes up a really strong case for re-engagement.

Read more: CDP Buyers Guide

Machine learning audience

Further examples of popular audiences are Churn Prediction and Customer Lifetime Value prediction.

The churn prediction audience is the most advanced type since it automatically identifies possible churners based on a churn prediction algorithm. The algorithm is based on behavioral data collected from the webshop and other channels.

For example, when customers decrease their purchase frequency, they are dynamically placed in the corresponding audience, and flows can be automatically initiated in the marketing platform.

The Customer Lifetime Value prediction also uses past behavior to define an audience automatically.

Below you can see an example of an audience with customers that have a high lifetime value but a poor chance of being active. This audience can provide valuable knowledge for the user and tell them if previous top customers are disappearing.