Логотип Help Portal
  • Начало работы

    Introduction

    • Onboarding overview
    • Project Initiation
    • Email compliance
    • Управление учетной записью

    Данные

    • Oбмена данными
    • Aдаптации данных
    • Web data collection
    • Интеграция

    Channels

    • Email onboarding
    • Smart Insight
    • Predict
  • Обучающие видео

    Resources

    • Introduction
    • Online Self-learning
    • Начало работы: видео

    Events

    • Instructor-led Training
    • Webinars
    • Seminars
    • Training Calendar
  • Руководства пользователя

    Стратегия

    • Персонализация
    • Автоматизация
    • Data Monitoring

    Channels

    • Email
    • Мобильные технологии
    • Web
    • Ads
    • SMS

    Add-ons

    • Smart Insight
    • Predict
    • Рекомендации по вознаграждениям
    • AIM
    • Relational Data
  • Статьи поддержки
    Forrester Wave CCCM (Independent Platforms) Report Q4 2019

    Новости

    • Что нового в платформе Emarsys?
    • Updates from the CSA
    • Пилотные функции
    • Статьи о защите конфиденциальности данных и GDPR

    Support

    • Как я могу получить помощь?
    • Редактирование профиля пользователя
    • Preparing for Black Friday 2019
    • Black Friday 2019 - Best practices
  • |
  • Partners

    Enhance Partners

    • Getting Started as an Emarsys Partner
    • The Emarsys Integration Platform
    • Automation Center Integrations
  • Разработчики
  • Статус системы
Нужна помощь?
Русский Deutsch English Español Français Türkçe 简体中文 Test New Chat
Войти
  • Predict
  • Web Recommender
  • 0 Predict Web Recommender

Содержимое раздела:

  • Overview:: Web Recommender - Overview
  • Overview:: Web Recommender logics
  • End-user guides:: Web Recommender templates
  • End-user guides:: A/B testing web recommendations
  • End-user guides:: Web Recommender for localized websites
  • End-user guides:: Web Recommender reporting
  • Uses cases and examples:: Web Recommender - Best practices and examples
  • Overview:: Mobile recommendations
EDIT
Expand all

Uses cases and examples:: Web Recommender - Best practices and examples

Updated: 13/11/2019 01:00
115004673225
Uses cases and examples:: Web Recommender - Best practices and examples

Before you start to design and implement your recommendation widgets, you should read these pages to get a better understanding of what Predict can do for you and your business.

Here we describe what we regard as the absolute minimum best practices that you should follow to take proper advantage of the automated revenue generation Predict can provide.

The Web Recommender widgets can be placed in any page on your website. They can display a variety of cross- and up-sell recommendations in different locations, in order to make the most of the customer’s visit to your site. Below are some examples showcasing these widgets.

 

Contents

  • How many recommendations to show
  • Using the Home Widget
  • Using the Personal Widget
  • Using the Related Widget
  • Using the Also-Bought Widget
  • Using the Cart Widget
  • Using the Category Widget

How many recommendations to show

Our recommendation models try to collect a minimum of 30 and a maximum of 100 product recommendations. 

The maximum number of recommendations that can be displayed in a web widget is 100. 

The number of widgets that you show, and the number that you make available via scrolling, will depend on your products and the page you select; just remember that the products are listed in order of relevance.

1. Using the Home Widget

This is a special widget designed purely for use on the home page of your web shop. Each visiting customer will see a different selection of products based on what they are most likely to be interested in, and displayed in sections designed to catch their interest and optimize conversion. This widget is aimed at both web stickiness, by providing an ever-changing shopping experience, and conversion, by recommending products right at the start of the session.

  • Click here for the recommender logic and code example.

The Predictive Homepage

The Predictive Homepage is a special solution which uses the Home widget to let you create a unique homepage for every visitor. You can give each and every customer a more personal and more engaging experience from the moment they arrive at your site. They will see recommendations personalized across categories and products, all delivered in the optimal order.

Even first-time visitors will get an optimized version of the homepage, as the products displayed will change in real time depending on the current browse and purchase behavior of your other customers. This means that you always promote the offers that have the highest chance of conversion at any given time. With this solution you can:

  • Dynamically tailor your homepage with relevant offers and recommendations.
  • Personalize across both categories and products.
  • Extend optimization to first-time, unknown visitors.
  • Easily integrate with any homepage design.
  • Create more sales opportunities with every customer visit.

How does the Home widget work?

You can place as many widgets as you like on your home page, and for each one Predict will select both the most appropriate category and the best items within that category.

This removes the limitation of having to stick to one category (if you were to use the Category widget) and allows you to personalize a far greater area of the page (than if you were to use the Personal widget).

Example

Imagine a customer is browsing your web shop, looking for different items for different reasons. Here’s what Predict thinks is currently of interest to them:

How do you make sense of this selection? By adding four Home widgets, each with six items, you might end up with something like this:

Now it’s easy to see what the customer is looking for. It helps them focus their search, and shows them that you really understand their needs.

2. Using the Personal Widget

The Personal widget offers targeted recommendations which match a visitor’s known characteristics and interests, based on their own browse history and the wisdom of the crowd. It helps them to find products they are interested in before they even know they want them.

  • Click here for the recommender logic and code example.

This widget is aimed at web stickiness and not conversion. It reduces browse abandonment by making it faster and easier to find the right product, and should always be prominently displayed on your home page as it gives an immediate personal touch to the browsing experience by displaying items that interested the visitor during their last session.

Using Personal on sub-pages

Since the Personal widget updates its recommendations with every click, you should also place it on generic pages such as Help and Support pages. This will ensure that the recommendations evolve with the session and adapt to the visitor’s changing behavior.

An ideal location is on the right side of the page, balancing the left navigation menu. If your design does not support this you can incorporate it into the body of the page, making sure that it is still clearly identified as a personal recommendation just for that visitor.

3. Using the Related Widget

he Related widget suggests alternative products to the items currently being viewed. It does this by looking at the browse behavior for the whole web shop and identifying product-to-product relationships. For example, if a visitor expresses an interest in cameras, the widget might display similar cameras as well as the current best-selling camera accessories.

  • Click here for the recommender logic and code example.

This widget is designed to boost conversion based on the customer’s browse history and the wisdom of the crowd.

The ideal location for this widget is right next to the product being viewed, within the product page. It should be placed either below or to the side of the product being viewed, so that the widget is visible without the need for scrolling.

4. Using the Also-Bought Widget

The classic ‘people who bought x, also bought y‘ widget, Also Bought offers recommendations based on items that have frequently been bought with the item currently being viewed. This is a more targeted version of the Related widget, and can be used well in combination with it.

It is particularly useful for web shops that sell items such as cameras which come with a lot of accessories.

  • Click here for the recommender logic and code example.

Located on product and cart pages, it can provide advice and insights, like an experienced shop assistant, improves the experience and enhancing loyalty.

This widget should be displayed once you are fairly confident that the visitor is preparing to purchase an item. For example, on the detailed description page of a product. It can also be used well in conjunction with the Related widget.

5. Using the Cart Widget

Similar to Also_Bought, the Cart widget offers recommendations based on the current selection of items in the customer’s shopping cart. It looks for relationships between these items and then makes additional cross- and up-sell recommendations (e.g. a spare power cord for cars, an extra SD card, etc.).

  • Click here for the recommender logic and code example.

This widget takes the recommendations one level deeper by analyzing the entire content of the cart before offering complementary items.

To maximize its impact, this widget should be placed in both the cart and the checkout pages of your web shop. On the cart page, it should be placed as close as possible to the actual contents of the cart; on the checkout page it should be located between the cart contents and the button to confirm the order and proceed to payment.  

6. Using the Category Widget

Similar to the Personal widget, the Category widget directs customers to the right products, but within a particular category, depending on the specific category page they are on.

For companies with a large and varied product catalog, it can direct the customer to the most relevant or profitable products within each category. As well as enhancing the experience with more targeted recommendations, it is very useful for inventory management.

  • Click here for the recommender logic and code example.

For websites with 100+ items in a single category, it is recommended to add the Category widget on the top of every category page. This immediately gives the viewer a personal highlight from within the category, reducing search time, frustration and increasing conversions and experience.

Была ли эта статья полезной?

Еще есть вопросы? Отправить запрос
В начало

Emarsys is a Leader

You may also be interested in:

Похожие статьи

  • Overview:: Web Recommender logics
  • Overview:: Web Recommender - Overview
  • End-user guides:: Web Recommender templates
  • Overview:: Mobile recommendations
  • Overview:: Predict - Overview
Copyright © 2019 Emarsys eMarketing Systems. All rights reserved
Legal Notice Privacy Policy Master Services Agreement Anti-spam Policy