Log in

DEVORAPRECIOS quadruples its conversion rate by implementing Retail Rocket’s AI personalization platform

DEVORAPRECIOS quadruples its conversion rate by implementing Retail Rocket’s AI personalization platform

DEVORAPRECIOS is an online store with a wide variety of household appliances, electronics and mobile brands. On their website, they define themselves as young entrepreneurs who, like their clients, look for the best prices. As their name indicates, their goal is to  “devour prices” through a lower commercial margin and by not charging unnecessary expenses.

Devoraprecios goals

Artificial Intelligence technology plays an essential role in e-commerce, and the Devoraprecios team knows it, this is why they contacted Retail Rocket to test the AI effectiveness in their then-recently built online store. 

The main objective was to carry out a comprehensive personalization to improve their e-commerce metrics. Great results have been achieved since the implementation, proving the huge potential of Big Data when you know how to use it.

Retail Rocket Solution

The consumer of household appliances and technology seems to be very well-informed. According to a report on online shopping in Spain in 2021, 85.2% of Internet users seek information before purchasing an electronics product, and 75.1% do so to buy household appliances. 

When looking for information about online shopping, most Internet users focus their searches on prices (89%) and information about the characteristics of the products (79%).

When looking for new appliances and technology, searching and viewing different options are essential stages in the purchasing process: these items are meant to last and they can cost quite a lot of money, so buyers generally prefer to take time and do some investigation rather than buy the first product they see. Here is where Retail Rocket technology plays an important role. The main challenge is to facilitate the visitor’s decision-making by showing a personalised online store for each user with product recommendations based on real-time personal interests. That will allow them to compare similar goods without leaving the website, and to find what they are looking for, shortening the visitor’s journey in the online store, and also increasing the conversion.  

Devoraprecios’ online shop is an excellent example of an integral web personalization in which the visitor will find recommendation blocks wherever they navigate within the online store. Therefore, AI technology covers the entire visitor’s journey in the purchase process.

To achieve the fixed goals, our team carried out the implementation of our platform on the website, and with the automation of triggered emails for different scenarios. 

The personal product recommendations are shown on the following pages:

  • Home page
  • Menu pages on the header and footer:
    • Brands page
    • Novelties page
    • Offers page
    • About us page
    • Contact form page
  • Drop-down horizontal category menu 
  • Category page
  • Product page
  • Checkout page
  • Exit-intent bar

For the exit-intent bar, automated trigger-based emails are sent for 4 different scenarios:

  • Abandoned cart
  • Abandoned product view
  • Abandoned search
  • Abandoned not available product view

The product displays are shown on a single line with horizontal scrolling in all recommendation blocks. In this way, the widgets are perfectly integrated on all the pages of the site. 

Let’s see in detail below the different types of blocks generated on each page while collecting and processing user information based on their real-time behaviour. 

Home Page personalization

The home page is where the customer journey on the web begins, and they forged their first impression that will lead the visitor to finish or not the sales funnel. 

Devoraprecios Home Page shows two main recommendation blocks: one with highlighted products on the web, and once the user’s information is obtained, with personal recommendations, and another one with the bestsellers. 

Here is a good case of a website with a high degree of personalization and dynamic content. The only fixed information for all visitors is the navigation menu in both the header and the footer. However, each user will see a different version of the online store according to their interests on each menu’s page.

In this way, we also find a selection of recommendations, for example, on the following pages whose accesses are on the navigation menus: novelties, brands, offers, items comparators, about us, account information, and the contact form pages.

Personalization on the horizontal category menu

In an e-commerce store with such a wide variety of items, easy access to each category is very important to keep the visitor surfing the site. The more accessible the products’ information, the easier to find what the user wants and, thus, to finish the purchase. And this happens especially when this kind of item goes together with other home appliances purchases. 

That is why the categories’ drop-down menu is fixed content in a privileged and visible place for the visitor on all the pages of the site. And of course, personalized blocks with recommendations are also generated within each category 

Personalization on the Category Page

The most popular products for each category are generated for this page based on the user’s interests and behaviour in real-time.

The menu with the categories displayed on all pages allows the user to move from one to another more easily. That increases the possibility of boosting the amount of products per order, especially for those customers who need, for example, to equip a new home.

Personalization on the Product Page

Once the visitor reaches this page, the chances of becoming a buyer are high. And in order for this to happen, each of the three recommendation blocks has its function.

To show similar products allows the user to consider other alternatives to the product they are viewing. In this way, if the price or characteristics do not fit with what the visitor wants, other options will be shown so that they do not leave the web.  

While the role of the first block is to keep the visitor on the page to find what they are looking for, the other two ones that contain related products and accessories are intended to increase the number of products in the cart. 

Personalization on the Checkout Page

Even if a user reaches the cart to complete the purchase process, we should not miss the opportunity to increase the average ticket per order. So getting the customer to add more items to the cart is the function of the recommendation block on this page.

Personalization on the Exit-intent Bar 

Several reasons can lead a visitor to leave the web: sometimes due to lack of time and others because the item they want is not available are just two examples. However, whatever the cause, the results obtained demonstrate the effectiveness of our platform to achieve the conversion of these users who, if it were not for the AI technology, would be lost customers.

When a non-registered user, of whom we already have information about their interests, is about to leave the website, a smart-form will be opened on the header. This form contains a product selection based on their behaviour and a CTA button through which to obtain their contact email address. It is useful both for the customer and the retailer, converting up to 4% of the visitors into subscribers while providing them with thoughtful and personalized service. This step is essential in order to cover the entire customer journey by automating newsletters for different scenarios, as we will see below.

Personalization of trigger-based emails 

Covering the entire customer journey would not be possible without the automation of trigger-based emails. In other words, implementing a single platform that integrates a comprehensive personalization is essential to achieving customer retention and improving metrics. Only, thus, by covering all the user stages, the retailer will obtain optimal results.

Collecting the user information based on their website behaviour is just as important as knowing how to treat the data, not only for web personalization but also for sending newsletters. Email marketing is a very potent tool both for customer recovery and for building up the loyalty of those who have already bought. 

Retail Rocket’s automated email platform is designed to deliver relevant and personalized newsletters to each customer in real-time. In addition, our smart-frequency-limit algorithm will not let email marketing become annoying. 

Retail Rocket triggered emails campaigns are designed to be implemented fast and easy. Sales and results will be generated without the need for additional attention.

Through the exit-intent bar of Devoraprecios.com, four scenarios of trigger-based emails are generated:

  •     Abandoned cart
  •     Abandoned product view
  •     Abandoned search
  •     Not available product view

Here is below an example of newsletters sent by the online shop:


According to the results achieved by using recommendation blocks, the effects of the personalization algorithms are higher while similar products are shown. As we mentioned at the beginning of this case study, this reaffirms that the consumer profile we are facing is a well-informed buyer who tends to compare different alternatives. The exit-intent bar and popular products within each category are the second and the third ones in the effectiveness ranking. 13.1% of the revenue comes from the recovery of visitors that tried to leave the web. If it is not for our platform implementation, they would be lost customers. 

Effects of recommendations

Without recommendations
With recommendations

Effects of trigger-based email campaigns


Effects of  email marketing personalization


  • Retail Rocket’s personalization platform increased online shop revenue by 21.45%.
  • The conversion rate with the recommendation blocks is four times higher than the conversion rate on the web with no personalization, meaning an increase in orders of 282%.
  • The average order value with Retail Rocket’s Artificial Intelligence technology is 4.16% higher with site recommendations and 36.58% higher with email marketing campaigns.

Devoraprecios & Retail Rocket testimonials

Thanks to the Retail Rocket technology based on Artificial Intelligence, in Devoraprecios, we personalize the entire Customer Journey. Integrating recommendation blocks on all website pages and emails allows every user to find what they need easily. In an online store with such a wide assortment, facilitating the search with a personalized offer is essential to offer the best service. 

Devoraprecios: David Conde, CEO at Principiar 

Working closely with the Devoraprecios team, thanks to the fluid communication with them, has been the key to increasing the online store’s conversion with our AI technology. We have personalized most of the website pages, and we are especially satisfied with the results obtained with the recommendation blocks on the product pages.

As for trigger-based emails, we want to highlight the excellent performance of the Price Drop scenarios, which show that communication at the right timing and channel is essential to increase metrics.”

Retail Rocket: Daniel Güidi, Business Development Manager  

Previous post

Next Loyalty 2022, let's walk together towards loyalty: 4-tips to make your customers happy with a personalised offer

Next post

What causes online shopping returns and how to avoid them through personalization [eBook]

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.