Fill the shopping cart with a recommendation engine

epoq Inspire for Inspiration in the web shop.

Did you know that…

  • 6

    More than 1 in 6 online purchases are impulsive? (Source: ECC)

  • 13,9%

    of online shoppers say that they were encouraged to make a purchase through online advertising? (Source: ECC)

  • 64%

    of shop visitors consider it important that businesses address them with relevant recommendations? (Source: Monetate)

  • 56%

    of customers are more likely to return to a page that recommends products? (Source: e-tailment)

Relevant recommendations - the sourceof inspiration to keep on browsing

A Recommendation Engine inspires online shoppers with relevant recommendations along the digital customer journey. For this, the click and purchase behavior of each shop visitor is recorded in a knowledge base and processed to a history. With a self-learning technology, each online shopper will be presented with customized products. The Recommendation Engine thereby stimulates the imagination of the online shoppers and enables them to recognize the product’s advantages. The emerging emotions lead to purchase impulses, which fill the shopping cart. That’s what epoq Inspire offers you.

Functionality of the recommendation engine
Functionality of the recommendation Engine

This is how it works:

    Shop visitors want to be inspired from the very first page – this is particularly important for fashion shops. It is important to meet the individual taste of the visitors and to inspire them when they enter the shop. The visitor is not always known to the site – in this case, the top sellers can help, or auto-generated recommendations based on general behavior across all visitors.

     

    If the visitor has already been to the shop, their click and purchase history is used to make relevant recommendations.

    Through the customer journey different touch points (points of interaction) can be personalized to talk personally to the online shopper. Relevant recommendations can be placed on different pages:

    • Welcome page
    • Category page
    • Product page
    • Shopping cart

    Thereby you offer a unique shopping experience to your customers. At every touch point, you can present your customers exactly those products that are relevant to them.

    Category or industry-specific requirements can be implemented through intelligent rules, so that the most relevant products can be recommended to a customer in the current context.

     

    Like in the drugstore area – if a user is interested in a shampoo for greasy hair, they are very probably a good candidate for other products in the range for their particular hair type.

     

    Najoba.de is able to offer personally-tailored recommendations to its customers according to hair and skin types, using individually programmed rules.

    “Often purchased together” – unite what belongs together. Recognize and actively offer thematically-linked products from your range: This creates a requirement and inspires the visitor.

     

    Like books, films and music – Harry Potter fans don’t just buy one book, but a second one too or maybe a Blu-ray, etc.

     

    Such targeted recommendations can be made right on the product detail page to encourage browsing. These are based on visitor behavior and customer history.

    “Other customers also purchased” and “recommended with this product” – find the right product and then buy everything at once. This is of particular importance in the area of fashion.

     

    Like men’s fashion – a luxury male outfitter offers his customers alternatives and knows which ties go best with which shirt.

     

    Generate personalized buying impulses, and offer good advice on matching outfits at the same time.

    Online Shops usingour Recommendation Engine

    The image shows the logo of the online shop ABOUT YOU. ABOUT YOU is one of epoq's customers.
    The image shows the logo of the online shop ALTERNATE. ALTERNATE is one of epoq's customers.
    The image shows the logo of the online shop babymarkt. babymarkt is one of epoq's customers.
    The image shows the logo of the online shop fahrrad.de. fahrrad.de is one of epoq's customers.
    The image shows the logo of the online shop BOGNER. BOGNER is one of epoq's customers.
    The image shows the logo of the online shop WorldShop Lufthansa. WorldShop Lufthansa is one of epoq's customers.
    The image shows the logo of the online shop BABOR. BABOR is one of epoq's customers.
    The image shows the logo of the online shop iba Buero Services. iba Buero Services is one of epoq's customers.
    The image shows the logo of the online shop JOCHEN SCHWEIZER. JOCHEN SCHWEIZER is one of epoq's customers.

    sister surprise Logo für Referenz der Welt Inspirieren durch personalisierte Empfehlungen

    SISTER SURPRISE EXCITES WITH INSPIRING LINGERIE AND SWIMWEAR:

    „The innovative Recommendation Service from epoq enables us to give our customers even more individualized and targeted product recommendations in our online shop and our newsletters.“

    Thorben Siewert | Head of Operations | vStores Commerce GmbH

    Leave a lasting impressionon your customers and visitors

    It is about matching visitors’ and customers’ taste with relevant recommendations and thereby creating a need. This is achieved by:

    • the right data models, such as these pants fit to this shirt, can be played out. Therefore, the algorithm makes use of the collected data in the epoq platform (knowledge base) and the specialized and industry knowledge of the shop owner.
    • testing new hypotheses through A/B tests and incorporating gained insights into the Recommendation Engine. Our algorithms learn independently, but can be controlled by our Data Scientist.
    • applying individual rules via the epoq Control Desk.

    We take on all the requirements as well as the specialized and industry knowledge of a shopkeeper and provide a tailor-made recommendation engine.

    Top-quality recommendations contain:

    DATA MODELS

    The models which provide the connection between data (e.g. these trousers are bought with this t-shirt) are controlled by the click and buy behavior of customers and visitors to the shop, as well as by human insight (hybrid approach).

    HYPOTHESES

    Hypotheses about the purchasing behavior of customers are first proposed and then tested with an appropriate method, such as an A/B test. The knowledge gained is then used in the recommendations.

    RULES

    Use-case oriented and highly-customized rules narrow the system’s recommendations down to the customer’s needs and the shop owner’s specialized knowledge of the industry.

    Integration?There's nothing easier than that!

    Product catalog

    In order to set up the Recommendation Engine, we first need your product catalog in XML or CSV format. We give you the tracking code for the connection to our platform, which you integrate into the shop.

    Tracking Code & Code snippet

    You add our tracking code to your online shop to connect to our platform as well as a code snippet as placeholder for the frontend.

    Configuration

    The relevant recommendations are configured in your test system until the optimal result is achieved.

    Go-live

    After a short learning phase, the relevant recommendations are displayed in the shop.

    Quality assurance

    We test the recommendations, adjust the rules to the desired strategy and provide training for the login.

    The Recommendation Engineis suitable for you, if

    • your online shop or product feed is running already and you want to optimize it
    • you have more than 1000 orders per month
    • you offer more than 1000 different products
    • you want to create an emotional experience for your customers

    You may also be interested in this case study

    Fully automated personalization in the online store leads to more sales. This increased the desired KPIs at Sister Surprise.

    Download Case Study Sister Surprise

    A / B testing shows differences between personalized and manual recommendations.

    Start typing and press Enter to search