Case Study – How Black Red White (Furniture&decoration) is able to enhance the shopping experience for their customers

For more than 25 years, Black Red White has had the most extensive offer for furniture and internal decoration furnishings available on the market. These include high quality furniture which is distinguished by its functional features, advanced technological solutions as well as attractive and diverse designs. Alongside the rich offer for furniture, BRW offers furnishings and accessories that are an indispensable part of the bathroom, kitchen, dressing room or dining room.

Black Red White is the largest furniture manufacturing group in Poland which sells 30% of its products abroad. BRW carries a wide array of products made available abroad in more than 30 countries worldwide, including the Czech Republic, Russia, Germany, France, Spain, Greece, Great Britain, Ireland and the Scandinavian countries.

The company has successfully developed its online store,, which records more than 2 million hits per month. The e-store plays two roles — selling and image. When it comes to shopping, the key factor which influences customers is the level of convenience. Therefore, the BRW brand seeks out a solution which is able to facilitate the process of searching for products as well as the purchasing process.


Recommendations generate almost 18% of the sales

One of the key elements of our sales strategy is to enable our customers to find the products which suit their individual needs as fast as possible and to encourage them to buy other products as well. We want our webstore to resemble a good sales assistant who understands the needs of our customers and recommends a personalized offer for every single one of them. – Magdalena Grzegorczyk, E-Commerce Manager at Black Red White.


We have managed to achieve our targeted goals with the implementation of QuarticON, a professional engine for personalized recommendations.


The QuarticON recommendation engine collects and processes data about the customers’ behavior and browsing patterns on the store’s website. Through our advanced learning algorithms (machine learning), the system builds a knowledge database comprising of the needs and preferences of each user as an individual and displays a personalized offer in real time. In this way, every customer can see a slightly different version of the webstore with products  carefully selected by our recommendation engine.


The store places more than 10 widgets with recommendations which generate almost 18% of the sales of the whole store per month. Depending on the placement of the widgets, each has a different strategy of recommending products whose objective is to increase conversion or  the value of the shopping cart.

The rise in conversion is influenced by the strategies which help customers find the products they need instantly. On the other hand, the increase in the shopping cart size is also influenced by the strategies which create new needs and cross-selling recommendations which help customers find and buy additional products.


Customer experience with personalized recommendations

A customer who enters a store has a definite need and idea of a product which he would like to buy. He passes on the information for our recommendation system about his needs by his behavior and initial browsing of the products. The recommendation engine learns the current needs of the customer from the very first click and responds to them by personalized product recommendations. At a predetermined place, the system displays the dedicated products for each user. The frame with personalized recommendations is placed on the main page, category page as well as on the ‘no search’ page and in an empty shopping cart. As a result, each user is able to see a customized offer and can find the product he needs instantly. For hundreds of thousands of users, we have achieved the same number of different versions of the store.


As compared with the bestsellers frequently used on the main page, personalized recommendations can contribute a conversion which is 10 times more from the same placement on the webpage.

Strategies based on personalized recommendations generate 8% of the monthly sales of the entire store on average.


The recommendations on the product pages which uses the browsing patterns of other users who are looking for a specific product also minimizes the time of searching. The product page includes recommendations like ‘Products that you may be interested in’. The learning algorithms analyzes the products which the customer considers simultaneously with the currently viewed product prior to making a purchase. The recommended positions are sorted automatically according to the index which determines the chances for purchasing a given product. Thanks to that, the process of searching for similar products is facilitated because the system shortlists only those items which have been regarded as similar by other customers. Such a recommendation strategy generates 5% of the monthly sales on average.


Immediately after the customer has chosen a product, which he wants to buy, the process of cross-selling begins, in other words, creating a new need for the customer and fulfilling that need. According to the interests of other users, the recommendation algorithm analyzes the additional products which have a higher possibility of capturing the interest of a particular customer based on the products which were selected or added to the cart by the customer. The product webpage shows complementary products for the customer — ‘Customers who viewed this item also bought’. For example, when a customer who wants to buy a bed, the system recommends mattresses and under-bed storage drawers. The recommendations are generated automatically in the order indicated by the rating which determines the highest chance of purchase. The order of showing products is very important because a statistical user usually clicks on the products in the recommendation widget starting from the left. By showing the same products which are in a different order, it tends to drastically decreased the sales from a given recommendation widget.


cross_selling_cartCross-selling recommendations are best noticed on a popup window which appears immediately after adding a product to the cart. As the customer is most focused on the choice he has just made, the additional recommended products based on the data analysis can rouse their purchase impulse. It has been proven by the fact that for every eighth customer who clicks on such recommendations (CTR on the level of 12%), and then every tenth decides to buy the recommended product (10% conversion)




Share this post with your friends

About the author

Paweł Wyborski

Leave a reply

Most popular

DYNAMIC VS. STATIC – what’s the difference?
18539497_l — kopia
Defining programmatic marketing and its 5 main benefits
6 Voice of Customer Tools You Should Be Using Now