Log in

How to achieve over 16% order increase with a personal approach by using Retail Rocket’s AI platform: Siberian Wellness case study

How to achieve over 16% order increase with a personal approach by using Retail Rocket’s AI platform: Siberian Wellness case study

The brands’ sustainability, both in terms of products and the company’s policy in general, is becoming more and more important to customers. Typically, these companies have a solid emotional connection with buyers and a high customer loyalty. What can brands offer to strengthen that bond while increasing the business’s KPIs? Let’s see how Siberian Wellness did it with an individual approach based on artificial intelligence

Figures and facts

Siberian Wellness is an eco-friendly Siberian brand with more than 24 years of history. It has retail outlets in 26 countries, an online shop available in 65 countries worldwide, and 2.5 million monthly online visitors.

The brand has a large community of fans from around the world. Since every customer is unique, Siberian Wellness must provide them with a personal approach. 

Retail Rocket Solution

The solution was to implement AI technology on the site and personalise the product recommendations for every single user.


Siberian Wellness goals

“We were born in Siberia, where we have been producing natural health and beauty products and sports supplements made of wild Siberian herbs for more than 24 years,” as the company states on the Siberian Wellness website. 

The company has a wide assortment, including eco-cosmetics, food and beauty products for healthy lifestyles, designed to meet different audience segments’ needs.

The desire to improve KPIs in a highly competitive market of “green” products was the reason to improve the service’s efficiency on the site through a personal approach to every single client. For this purpose, the following goals were set:

  • To improve the service by personalising product offers in the online store;
  • To increase the revenue and the average order value by displaying, on every site’s page, personal product recommendation blocks based on the user’s interests in real-time;
  • To increase and build customer loyalty.

“Our brand is well-known not only in Russia but also beyond its borders. And despite strong customer relationships, there is always the opportunity to improve the service to make the online shop’s customer journey easier and increase metrics. Retail Rocket proposed the personalisation of product recommendations shown based on a sophisticated mathematical model. Thanks to their technology’s integration, every single customer can quickly and effortlessly find the suitable products and easier navigate through the Siberian Wellness assortment, which increases the average order value and orders” – Daria Veligodskaya, E-Commerce Director, Siberian Wellness.

Retail Rocket Solution

Organic food and cosmetics fans have different shopping approaches when paying attention to the product composition, price and other characteristics. Thanks to the Retail Rocket platform’s Data Warehouse module, all customer data is stored in a single source: his preferences, interests, behaviour on the site, purchase history… which allows segmenting the target audience, offering every customer a relevant product selection at every stage of the Customer Journey. Moreover, the artificial intelligence – the basis of the Retention Management Platform modules – knows what assortment of products to offer the client at any given time. Taking into account the specifics of Siberian Wellness, Retail Rocket specialists designed the following strategy:

  • To provide every customer with a unique version of the online store, based on their needs and interests;
  • To implement personal product blocks on the online store’s main pages;
  • To increase the product recommendations’s effectiveness on site by using  A/B testing technology and the Growth Hacking team’s expertise.

We analysed the Siberian Wellness online store’s functioning, collected customer data, developed a personalisation strategy for every page, and then conducted A/B tests of hypotheses, highlighting the most effective interaction mechanisms with customers. All of this had a positive impact on business metrics, ensuring their growth” – Dmitry, Retail Rocket Manager.

We created a personal version of the site for every single customer 

An individual approach for every customer increases their loyalty, creates a positive shopping experience and helps convert a new customer into a regular one. When a website is adjusted in real-time to the visitor’s needs, they will easily find what they are looking for since Artificial Intelligence will consider parameters such as products and categories of interest, price, size, colour, season, etc. 

Retail Rocket has designed and implemented personal recommendation blocks on 15 key pages of the Siberian Wellness online store: on the home page, category page, product page, shopping cart, promotions page, blog, search page, etc. Let’s look at the personalisation on some of them:

We captured the visitor’s attention with the best offer on the home page

The home page, like a showcase, is designed to display the most exciting offers and immediately attract visitors’ attention. Siberian Wellness shows a product selection to attract both new and regular customers.

While new users will be able to familiarise themselves with the store’s bestsellers, those who have already bought will be shown personal product recommendations that might be of their interests based on the information stored with their purchase history.

We make it easy to search by category with a personalised offer for every user

It is unlikely that a visitor interested in a specific category will want to see all its products. Therefore, it is essential to display a product offer selection based on their interest on this page.

New users, whose preferences are not known yet, will see the bestsellers within the category. Those customers who have already interacted on the site will be shown personal recommendations based on their interests and behaviour.

The intelligent personalisation algorithm in the main category will also display personalised recommendations for all its subcategories. For example, in the “Beauty” section, a customer will see a selection not only of the body creams they have been browsing but also of cosmetics and perfumes that might fit their interests.

We delight the visitors with personal discounted products

On the promotion page, Siberian Wellness shows its customers discounted bestsellers. This increases the likelihood that the users add the products they like to the cart  and complete the purchase. 

We help the user to make a purchase decision on the product page

When a visitor enters a product page and shows a clear interest in a specific product in the catalogue, they are one step closer to completing an order. Besides providing detailed information about the product at this sales funnel’s stage, displaying a recommendation block with similar offers is another excellent option to help users find what they are looking for and make a buying decision.

Personal recommendations on the website’s blog

Besides articles of interest for Siberian Wellness visitors, a recommendation block with bestsellers and novelties is displayed on their blog. They will be changing in real-time based on the post’s section: beauty, health, nutrition, etc. 

How we increased the AI personalisation’s efficiency by using the A/B test mechanics

We conducted nine A/B tests that allowed the retailer to offer every visitor the best products based on their interests and behaviour, increasing the AI technology’s effectiveness.

Let’s see how to continuously boost the metrics of your online store with these mechanics through iterative improvements of Siberian Wellness. 

Personal offers in the category

In online stores with a wide assortment of products and categories, it is not always easy for users to browse and find what they want to buy. That is why we implemented a personal recommendation block and a selection of bestsellers in the category, saving their time and shortening the customer journey. 

Let’s see what led us to make this decision.

To select the most efficient algorithm, our Growth Hacking specialists conducted several A/B tests. All site visitors were randomly divided into three segments:

The first segment was shown personal product recommendations in the category:

The second segment was shown the bestsellers in the category:

The third segment, the control group, was not shown any recommendations 

The test showed the following results:

SegmentConversion Average order ValueRevenue
Personal recommendations in the category+12.28%-3.75%+8.07%
Bestsellers in the category+5.22%+0.26%+5.50%
Without recommendations

Personal product recommendations in the category showed the best results with a statistical significance of 97.4%: a 12.3% increase in conversions and 8% growth in revenue.

Product recommendations from other categories in the cart

When a customer adds a product to the shopping cart, personal recommendations are also very effective. Therefore, at Siberian Wellness, we conducted A/B tests to determine what configuration offers the best results. 

Site visitors were randomly divided into three segments.

The first segment was shown related products from categories other than the product in the cart.

The second segment was shown related products:

The third segment, which was the control group, was not shown any recommendations.

The test achieved the following results:

SegmentConversion Average order valueRevenue
Related products from different categories to the product in the cart+7.35%+4.33%+12.00%
Related products +2,70%+2.16%+4.92%
Without recommendations

The best results were achieved by the first segment, “Related products from categories other than the category of the product viewed in the cart”: conversion growth by 7.35%, average order value growth by 4.3% and revenue increase by 12%. With a statistical significance of 99.3%.

Improving search recommendations

Those customers who know the category or product they want to buy are likely to look directly for it using the search bar on the site. It is, therefore, essential to implement Artificial Intelligence to help the visitor in this stage.  

Conducting an A/B test led us to select the best option for the search page personalisation.

Site visitors were randomly divided into two segments:

The first segment was shown search recommendations:

The second segment, the control group, was not shown any recommendations.

Test results:

SegmentConversion Average order ValueRevenue
Search recommendations-3,24%+6,75%+3,29%
Control group

Search recommendations showed an increase in average order value of 6.75% and generated a 3.3% increase in revenue.

The hypothesis with the highest increase in KPIs was chosen based on the results obtained in the conducted A/B tests.

In this case, there was no significant difference in conversion rates, and the test results would have been dismissed if only this metric was taken into account. However, a significant increase in the average order value made us consider the hypothesis. 

Our main goal is not only to test a hypothesis but also to analyse different site’s sections to find the most effective and valuable solution for the business (in this particular case, to provide an increase in revenue due to a significant growth in the average order value).


By using an intelligent online merchandising strategy and increasing efficiency through hypothesis testing by the Growth Hacker team, we managed to achieve the following results:

  • Personal recommendation blocks generate 16.2% of orders on the site;
  • Conversion in recommendation blocks is 13 times higher than the average on the site. 


Our eco-brand offers customers products of different categories: from healthy food and drinks to wellness products and clothing, and it is essential, for us, to understand the needs and guess every customer’s desires. Personal product selections on different pages of our site save customers time and allow them to see the maximum of interesting offers from our assortment. As a result, everyone wins: business KPIs grow, and customers are satisfied with our service and products. Thanks to the Retail Rocket team for in-depth analysis, recommendations and effective solutions

Daria Veligorodskaya, Ecommerce Director, Siberian Wellness.

“Siberian Wellness has won the love and trust of millions of customers around the world. In a business developed offline and online, we found points to improve the site’s key metrics. The excellent and constant communication with the Siberian Wellness team and their desire to develop using IT trends helped us implement a strategy that ensures the growth of average order value and revenue per customer.”

Dmitry, Retail Rocket manager.

Previous post

How to easily automate trigger-based emails and quickly increase your revenue

Next post

Churn Rate features: why customers leave you and how to avoid it

Did you like the article? Subscribe to the newsletter to receive fresh articles in the mail.

Subscribe to the newsletter

We use our own and third-party cookies to obtain statistics of the navigation of our users and improve our services related to your preferences. You can configure your preferences. You can get more information here.