Ideen für Personalisierung im Onlineshop, GK Software SE

Have you ever been frustrated while shopping at a retailer’s online store? Under what conditions do you feel that you want to continue shopping?

Personally, I spend more time on a retailer’s site when I see products that match with my preferences and meet my taste. The truth is, the majority of online customers are like me. Intelligent personalization in the online store is fundamentally linked to a great shopping experience. Studies1 on the topic of personalization have regularly shown that a positive shopping experience also helps online retailers increase customer loyalty and sales, and to reduce returns in the long term.

Online shops are always finding new ways to inspire and retain their visitors. Since almost 70% of online shoppers2 now also shop via smartphone, the optimal positioning of relevant products on a smaller screen is more important than ever. This is exactly what personalization in the online shop does: display the “right” products, tailored to the preferences of the mobile shopper.

In the years after the turn of the millennium, simple algorithms implemented recommendations à la “customers who bought this product, also buy these products,” but over time personalization made its mark in new and innovative ways in online shops, newsletters or apps. Intelligent systems now interact in real time with every single customer and show suitable products and content in seconds based on their click behavior.

Below you can read five ideas for intelligent personalization in the online store that will inspire your customers and boost your sales:

1. Successful personalization in the online store: Rely on recommendations based on image similarity

It often happens that a product is liked and added to an online basket, but not purchased. This may be because the customer is searching for a product that looks similar to the item displayed. Product recommendations based on image similarities can fulfill this customer need. AI algorithms analyze all product images and recognize similar shapes and colors. In this way, the customer receives product recommendations that are visually similar to the original product. If the customer looks at a green wristwatch, he will receive recommendations for wristwatches that look similar in terms of shape, design or color. The recommendations based on image similarities can also be filtered based on information in the product master data, for example, that only products from a certain category or a certain brand are recommended.

Image Similarity, Bildähnlichkeiten, Personalisierung, prudsys, GK Software SE

Recommendations based on Image Similarity

2. Align product recommendations with your KPIs and reduce the number of abandoned carts

Good personalization helps achieve your key KPIs. An effective feature for this is the linking of product recommendations to the price. If you want to increase shopping cart value, you can set the corresponding minimum price for the product recommendations. For example, you could determine that the product recommendations may be a maximum of 5% cheaper and a maximum of 40% more expensive than the displayed product. Or you can use the personalization solution to only recommend items that are in a certain price range.

Produktempfehlungen gekoppelt an Preis, prudsys, GK Software SE

Recommendaions in a certain price range

Another example: You want to increase your sales with a “daily deal.” In order to encourage as many visitors as possible to use the daily deal, it is best to align it with the preferences of the customers. If you show the customer an item from their favorite category or a product that is visually similar to the product they have just viewed as a daily deal, the chance that they will be interested is higher than with a more generic deal.

Another scenario: What is your average number of abandoned carts? The number varies by industry and company, but the Web-UX-Institute Baymard4 assumes an average of 70% of shoppers abandon their carts. One way of reducing the number of abandoned shopping carts is by offering incentives. For example, if a certain time window (or other criteria defined by you) for completing the purchase has been exceeded, an intelligent personalization solution can display a discount or free shipping costs for completing the shopping cart.

3. Promote cross-selling with dynamic recommendation boxes and a gamification effect

Making your customers aware of other topics and product categories in your online shop is essential. The majority of customers only move within their favorite categories. Inspire these customers by displaying additional product ranges and thus raise hidden sales potential.

This can be achieved by displaying dynamic recommendation boxes. Customers like it when they can determine what they see for themselves. For example, you could include recommendation boxes in which the customer can click on their favorite category and then see personal product recommendations. This increased interaction with the customer provides the personalization software with additional data for targeted personalization. You can also save space through the compressed display of the recommendations and can use this for further personalized areas (such as a keyword “daily deal” or “product bundles”).

Another feature for your customers: offer visitors who like to play games an incentive. Arouse their curiosity by presenting a hidden, personal recommendation as a “secret” favorite product. The customer can actually view the hidden personalized product with one click.

4. Achieve a high degree of personalization through the use of a digital customer card

The reliable customer card is far from obsolete. Digititally, the customer card appears in the form of a profile and is still a popular customer loyalty tool. Due to the large amount of information that personalization software receives through customer interactions,  precise product recommendations can be calculated.

Online shoppers are often willing to share their interests and preferences if it helps them get better recommendations or tailored deals. Customers now want to participate in personalization themselves3 by creating customer profiles. In fashion retail customer profiles can include, information on clothing style, preferred fit or favorite colors. Information about skin type, eye color or favorite colors can be noted in the beauty segment.
With the additional information applied to your product segment, you can further increase the quality of the recommendations, increase the satisfaction of your customers and ultimately generate more sales. You can also use this customer information for personalization in other customer channels such as newsletters or online shops.

5. Increase shopping cart value through the intelligent coupling of categories

When online customers go looking for gifts, inspiration is helpful. An example of this is the search for suitable birthday presents for a child. For example, the shopper looks for a piece of clothing in a certain size. How practical would it be to receive recommendations for toys that match the trousers you are looking at and that roughly correspond to the child’s age? This can be achieved by coupling the clothing size sub-category with the toys sub-category for a certain age group. Or you can use the shopper’s previous purchases to determine the age of the child. Based on the age, you can generate product suggestions in the regular newsletter that “grow” with the age of the child.

Test out which form of personalized customer approach works best in your online shop.

Would you like to implement one of these use cases or do you have your own ideas for exciting scenarios in terms of personalization in the online store? Talk to our personalization professionals who have already successfully implemented many projects for different customers. Make your appointment here.


  1. Evergage: “2019 Trends in Personalization”, under:
    https://www.evergage.com/wp-content/uploads/2019/04/2019_Trends_in_Personalization_Report.pdf (retrieved at 1-29-2021)
  2. Der Bank Blog: „Studie Paypal – Mehrheit der Deutschen shoppt und zahlt mobil“, under: https://www.der-bank-blog.de/mcommerce-deutschland/studien/37663510 (retrieved at 1-22-2021)
  3. IDG ChannelPartner: „Kunden wollen an Personalisierung teilhaben“, under: https://www.channelpartner.de/a/kunden-wollen-an-personalisierung-teilhaben,3334612#:~:text=Doch%20Studien%20zeigen%2C%20dass%20der,kann%20als%20f%C3%BCr%20die%20Firmen.&text=Unternehmen%20verstehen%20unter%20%22Personalisierung%22%20relevante%20Inhalte%20und%20eine%20pers%C3%B6nliche%20Ansprache.&text=sich%20an%20Kaufentscheidungen%20des%20Kunden,relevante%20Angebote%20machen (retrived at 01-27-2021)
  4. Web-UX-Institute Baymard: „44 Cart Abandomnment Rate Statistics”, under, https://baymard.com/lists/cart-abandonment-rate (retrived at 01-12-2021)