Mobile/
E-commerce/
Portals

Polish m-commerce keeps growing, but there’s room for improvement

 

This article was reviewed and updated on July 24th, 2023

Findings from the second edition of Google Polska and e-point SA’s survey on Polish mobile stores.

Is Polish m-commerce ready for important e-commerce events, such as Black Friday, Black Week, or Cyber Monday?

The results of the second edition of Google Polska and e-point survey indicate that Polish entrepreneurs find the mobile sales channel very important. Furthermore, they are willing to improve it. There are many areas where better results could be achieved; these improvements could prevent the loss of valuable sales opportunities, particularly in view of the anticipated recession.

M-commerce (mobile commerce) constitutes a significant portion of any e-commerce activity. This can be clearly seen in the growth figures and dynamics of the segment.

In 2011, smartphones accounted for 6% of global web traffic. In 2022, this share grew to over 55%.

Such an increase should not be ignored – especially because now more people own a mobile device than a desktop computer.

These global changes can also be observed in Google data. A few years ago, Google analysts noted that smartphones outstrip desktops in e-commerce. In 2021, smartphones were responsible for two times the number of requests made from desktops. According to the newest data obtained by Google (in 2022), Polish Internet users used the Google search engine 2.5 times more frequently on their smartphones than on traditional computers. This trend has been reflected in the structure of e-commerce traffic, where in some industries the proportion of mobile internet users amounts to 80%.

Competition on this market keeps getting more intense; only those best prepared will be able to succeed. This observation led to the project conducted by Google and e-point SA. The goal was to find out how to enhance Polish m-commerce, making it even faster and more friendly to internet users. The second edition of the project encompassed the audit of 25 mobile company websites, which was aimed at streamlining the sales process and improving the speed of mobile stores.

"The first edition [of the survey] drew a lot of attention from the industry. Therefore, we decided to continue the project under the Google Retail Growth program. The conducted audits helped entrepreneurs to better use the potential of m-commerce, allowing them to adapt their stores to the habits of mobile users and the potential of users’ devices – thus increasing the probability of closing a sale".

Piotr Kowalski

Senior Analytical Consultant

Google Polska

Linkedin logo

Company owners and e-commerce managers are well aware of how difficult it is to win new customers. They put in a lot of effort to manage paid traffic and advertising campaigns. They continuously strive to encourage users to visit the service, hoping this will boost conversion. However, a sluggish mobile service with a poorly designed interface may strongly discourage mobile users from making a purchase.

"Numerous reports seem to confirm that m-commerce is of paramount importance in the entire e-commerce portfolio. Thus, the findings of this survey – aimed at identifying room for improvement in the Polish m-commerce – constitute a great value for entrepreneurs who want to scale their businesses accordingly. Technical aspects are of great importance, as they enhance the functioning of mobile store versions and drive satisfactory levels of user experience in terms of speed. Whenever the number of customers visibly falls in the channel, attention should be paid to how services are designed and what technologies are chosen".

Piotr Różycki

Key Account Manager

e-point SA

Linkedin logo

Survey methodology and key findings

The mobile UX analysis used 58 checkpoints to establish a base for studying user journeys according to two scenarios. The website speed survey involved the analysis of speed indices known as Core Web Vitals (which describe how quickly elements show during the loading of a website, how quickly the site responds to user action, and the visual stability of website elements during loading). Based on these findings, companies received reports including detailed descriptions of proposed changes to enhance their mobile store’s UX and operating speeds.

Top 5 UX recommendations

Most mobile customers browse products or services using one hand; this often happens during breaks or between other activities. The customer’s attention is distracted by various activities and focuses only momentarily on a specific task. Therefore, m-commerce must strive for very high efficiency in properly accommodating customer needs. A list of 5 recommendations to enhance the mobile user experience is presented below.

  • 1. Allow adding to favorites without registration

    Being able to add products to favorites is a great benefit while browsing the product catalog – it lets the user save the product for later reference without adding it to the cart, which always implies a purchase intent. The "Add to favorites" option should also be available to clients who are not logged in. This allows the user to smoothly browse the store without registration, which may have a positive effect particularly during the first longer visit.

  • 2. Important actions should be visible above the fold

    Because of the small size of smartphone screens, all important information and actions should be clearly visible above the fold (i.e. immediately visible, without forcing the user to scroll) after the website opens. On a product site, this will be the product price and the "Add to cart" button. For a product category view, users should see at least a major part of the first product on the list.

  • 3. Provide the right keyboard for the task

    On touchscreen devices, users enter data using a keyboard displayed on the screen. Modern browsers adjust the displayed keyboard to the nature of the data entered – e.g. in the telephone number box, the presented keyboard should be numerical and have large digits; in the address box, an alphabetical keyboard is needed.

      

    Putting an alphabetical keyboard with small digit buttons in a number input field causes user frustration and increases the risk of entering a wrong telephone number, address, postal code, or PIN number. This may become the source of unnecessary problems and costs in order processing. It is highly recommended to adjust the type of screen keyboard to the expected contents of the form. This will make form completion a lot more friendly and error free.

  • 4. Errors should be indicated by more than color

    During the completion of the order form (i.e. immediately before checkout), errors should be clearly marked, with a distinct icon next to the incorrectly completed box and a clear description of the problem.

      

    Marking with color (e.g. turning an empty field red) may be a useful supplement and a way of drawing attention. Yet, on its own, it is not sufficient to clearly indicate the error. This is particularly vital to users who do not distinguish colors; they need an additional indication of the wrongly completed form boxes.

  • 5. Visible filter status

    Using filters is a fundamental way of searching relevant products in a category. In the course of one session, users may frequently switch their applied parameters. The filter setup should be visible to the user without any further interaction (e.g. opening the filter panel). Thanks to this, users can avoid mistakes and better understand the presented content.

Top 5 speed recommendations

The average smartphone has 6 times less screen than a desktop, 4 times slower processor speeds, and less RAM. These limitations are exacerbated by a slower internet connection or even no connection at all (e.g. while traveling).

Faced with the dynamically growing popularity of the m-commerce channel, we should stop treating mobile as a "second-class option". Instead, we should design and test it with the same diligence as the traditional desktop. Streamlining mobile store options is key to turning the browsing user into a customer. Thanks to the conducted audits, we have identified 5 recommendations that can enhance the operating speed of mobile stores:

  • 1. Review JavaScript

    If JavaScript blocks rendering, the browser must download, process, compile, and assess the script before it performs any other actions necessary to render the website. Even if JavaScript is asynchronous (i.e. does not block rendering), the code will compete for bandwidth with other resources while downloading, which significantly affects efficiency.

      

    Additionally, unused JavaScript may slow website download speed. Plus, sending unused code is wasteful for mobile phone users with limited data plans.

  • 2. Optimize the cache policy

    When a browser requests a resource, the providing server will tell the browser how long it may temporarily store or cache the resource. For each consecutive request of this resource, the browser will use its local copy instead of downloading it from the Web. Thus, HTTP cache optimization can significantly reduce the time of downloading during multiple visits.

  • 3. Upgrade graphics and images to next-gen formats

    WebP is an image format that has better compression and quality (as compared to the older equivalents of JPEG and PNG). Using newer formats means images will load faster and use less mobile data.

  • 4. Optimize the size and structure of the DOM tree

    A large DOM tree (Document Object Model, a common model of website structure) may in many ways slow down the website efficiency.

      

    In terms of the Web and capacity efficiency, a large DOM tree usually contains many nodes. These nodes are not visible when a user loads the site for the first time, but they contribute to unnecessarily high data costs and slow downloading times.

      

    This also affects the efficiency of the mobile website. The more interaction users and scripts have with the website, the more the browser must keep recompiling node positions and styles.

      

    A large DOM tree along with complex style rules may significantly slow down rendering. If the applied JavaScript uses general request selectors (such as document.querySelectorAll), the website may store references for a high number of nodes. This can overload mobile devices’ memory capacity, thus reducing memory efficiency.

  • 5. Review CSS styles, sizes, and service

    Unused CSS may also slow down the browser’s construction of the rendering tree. A rendering tree is similar to the DOM tree, but it also contains styles for each knot. To build a rendering tree, the browser must go through the entire DOM tree and check which CSS rules apply to each node. The more unused CSS there is, the more time the browser will potentially need to calculate styles.

      

    Good CSS file minification may improve the website loading efficiency. CSS files are usually larger than they should be.

Conclusion

  • 73% of companies that implemented at least one mobile UX recommendation saw increased conversion in that mobile channel (compared to desktop).*
  • 57% of companies that did not implement any of the mobile UX recommendations saw conversion in that mobile channel slump (compared to desktop).
  • Companies that implemented recommendations received an average increase of 12% in website speed; those that did not saw a 5% decrease. 
  • Over 530 recommendations on mobile user experience (UX) and portal speed were given to 25 large- and medium-sized Polish e-commerce companies. These companies operated in the categories of health and beauty, furniture, food products, consumer electronics, books, fashion, and tourism.
  • 72% of companies implemented recommendations; over 30% of these recommendations had been or will be implemented. The average audit assessment by the participating companies amounted to 4.9 points out of 5.

Footnotes:

*) The metric used in calculations was the Relative Conversion Rate (Mobile Conversion Rate and Desktop Conversion Rate quotient), which describes a relative probability of making a purchase in the mobile channel versus the desktop. Calculations were made using the 'pre-post' method, comparing the period of 4 weeks before the audit presentation against the period of 4 weeks at the end of the project.