Empower Your Business With The Magic Of Machine Learning!
2 out of 100 customers buy when enter your eshop and 20 out of 100 in store because they find it difficult to discover products they desire.
Retail Industry is going in the direction of customer personalization in e-commerce and in store.
Every e-shop from now on will offer a very personalized customer experience.
Every store will use Intelligent shopping assistants that communicate with incoming customers with speech , recommend personalized products on digital in store displays and will collect customer contact and shopping data.
In eshops each customer sees something completely different from the previous or next one.
Using Artificial Intelligence and Machine Learning Predict the products that each customer wants to buy individually and recommend them.
When he logs in we can understand who he is and the experience he will have will be really impressive.
So practically from now on when the customer enters your eshop you will have to convince him to log in.
He will be able to discover without any effort products that he had not imagined that you had in your eshop and even more impressive that he himself had not imagined that wanted them.
It was somewhere in his subconscious that he needed it to take his life a little further.
The technology of Recommendation engineering discovers for your customer products they hadn’t thought to search for.
Until now eshops have not tried to a large extent to make the user login because they had nothing impressive to offer , a significant benefit.
These will be different from customer to customer.
So when he logs in and logs in the Home page will be different from user to user.
When entering the product category the category will be different from user to user.
He will have products that he wants to buy.
When he clicks on a product below he will be shown Recommended products that he wants to buy.
When he puts products in his shopping cart, products that he wants to buy or are bought together with this product and he hadn’t thought about will appear underneath.
Not in general but specifically how your business works.
Don’t forget that recommended products are created by the recommendation engine by learning from your special customer purchase data.
They are products that your customers already buy or want together but until today you haven’t used them because you didn’t have Artificial Intelligence to discover them.
So below we will give some examples of how your eshop will be using recommendation engineering, offering an impressive user experience and increasing your revenues up to 30%.
The user will now be able to enter your eshop and discover products and services that make him happy without any effort of thinking.
You won’t need multiple emails every month to every user to catch the one who wants to buy something.
You’ll give him an exciting experience and he will come back again and again!
Consumers and all people don’t like to think too much.
They want to use thinking (system 2 based Kahneman ) for things that matter.
They want to be offered products and services that satisfy them and they simply choose them.(system 1 Kahneman )
We’ll use Amazon as an example here who knows recommendation engineering very well but there are thousands of others you can see like YouTube , Netflix , TikToc , Alibaba ,Facebook , Linkedin , Spotify etc.
It therefore offers a unique , special and amazing experiences to every customer.
Every customer finds products that interest him without much thought and effort and makes his life more beautiful.
Looking at Amazon’s design we see the following:
The recommendations are practically visible on every page, tools and devices taking endless shapes and forms.
In the mobile version of the eshop 45 different recommendation widgets were identified on the homepage alone.
These recommendation widgets present the products that the user will see using advance machine learning with several variables per user, such as:
・ Location
・ Recent purchases
・ Clicks per products , by category, shopping trends and seasonal trend
・ Stored items or lists by consumer
・ Offers or discounts
・ Purchases completed by other customers after viewing similar products
・ User Reviews
・ Top choices for customers
Both the eshop and the app mimic each other, and each features lengthy pages with dozens of unique recommendations.
Some recommendations promote product discovery, while others encourage users to buy previously purchased products.
Content recommendations often take the form of advertising banners and also appear in promotional messages.
In general, Amazon uses the homepage as one of the main beacons of discovery.
A unique use case of recommendations that Amazon is developing is on product detail pages (PDPs).
The recommendations, presented below, are almost unlimited.
Amazon does not limit the number of recommendations per widget on these pages and tailors the discovery experience across devices.
On the desktop, it lists the number of products per widget in the top right corner, presumably using its algorithm to display as many product recommendation as possible that match the predefined criteria for that widget.
On mobile, it deploys a horizontal infinite scrolling feature, allowing users to endlessly switch between products that are similar to the current product in view.
The fixed homepage menu on the desktop site is optimized for logged-in users , displaying more menu options , such as “Buy Again” and “Browsing History” as a push towards product discovery and creating a better user experience.
These menu items do not serve primary/anonymous users because they have no purchase or browsing history accessible.
Instead, they are served exclusively with the basic menu items
(Amazon.com,” “Today’s Deals,” etc.).
Encourage repeat purchases in the home page menu bar for known users
“Buy Again” is a purpose-built page that is extremely useful for Amazon users.
They are able to browse through previously purchased products with ease and repeat purchases, which is unusual in the world of e-commerce retailing.
Platform:Web, Application
On the homepage, Amazon recommends items based on your browsing/shopping history on the homepage, sorted by category.
Platform: mobile web, app
Amazon gives personal recommendations based on ongoing agreements that meet the user’s requirements,correlations and interests.
Platform: Desktop
Amazon offers personalized content recommendations for returning users (music, videos, content, etc.) and graphical recommendation elements based on previous user history.
Platform: Web and mobile
The home page on the mobile web includes a flow with an endless scroll, featuring several, personalized graphical elements.
This encourages users to review a recent purchase from their home page feed feed.
Platform: mobile web
After reviewing an exhaustive list of personalized recommendations on the home page, users are provided with a message explaining that this is the end of the initial proposal experience and are then encouraged to browse the product categories.
Users are served with a very specific type of recommendations in Buy Again.
Recommended products are those that other Amazon customers frequently buy again, as well as products that are “frequently repurchased” in specific categories.
Platform: Desktop
In the drop-down menu on the home page, users are served up to 3 products recommendations items they have purchased along with menu options, giving users easy access to repurchase these items (“Buy it again” section).
To learn how the top companies in the world increase revenue up to 30% from existing customers using Recommendation Systems and Personalization read the book below.
https://mobiplus.co/ebooks/customer-rediction/
Platform: desktop, Mobile Web, App
Within category and section pages, Amazon serves up category-specific recommendation images to all site visitors.
Amazon implements a number of recommendation strategies through these widgets, which include well-known offers and best sellers.
On both section pages and category pages, recommendations appear almost or completely below the box, and are stacked on top of each other, across all channels.
Platform: Desktop
Connected users are provided with product recommendations in the category and section.
Each recommendation component has a different strategy, such as
Platform: desktop, Mobile Web, App
Anonymous, new users receive recommendations for
“Best seller,”
“Most wanted for,” and
“Top rated” on category pages.
In the app, up to 6 recommendation widgets are displayed.
Up to 5 recommendation widgets are displayed on the desktop.
And on the mobile web, users can see up to 6 recommendation widgets.
Recommendations for all users on all channels
Platform: desktop, Mobile Web, App
Within the category pages, Amazon recommends products that have seasonal relevance, i.e. Halloween costumes on the Clothing section page, holiday gifts in many categories during the month of December, etc.
Platform: desktop, Mobile Web, App
On some PDPs, Amazon displays a widget below the fold with side-by-side comparisons of similar products to the product in question.
These widgets are compared upwards of 6 product variants, the first of which is the primary product.
Platform: desktop, Mobile Web, App
Amazon shows bunners on the pages following some PDPs in Amazon’s desktop eshop.
These banners promote Amazon gift cards and promote high-quality video content to both known and unknown users.
Some PDPs display third-party advertisements in place of an Amazon home banner.
Platform: desktop, Mobile Web, App
Amazon presents different offers in the PDP descriptions to promote loyalty programs, sponsorships, and more.
This method allows them to highlight different financing tools, such as Amazon Reward Signature Cards.
In this way, they try to integrate more paying users into their credit card programs, as well as offer discounts at the time of purchase.
Recommendations from popular products
Platform: desktop, Mobile Web, App
Users are provided with personalized recommendation within categories, by segment, popularity and trends for that category.
Platform: desktop, Mobile Web, App
Users are provided with recommendations based on their “shopping trends” -these are usually items they have not seen before and instead fill in products they may be interested in.
One of the first promotional offer encounters in the eshop is displayed within the top menu bar.
“Today’s Deals” directs all site visitors to a home page with selected products that are discounted especially on that particular day, along with a countdown counter in the description of each product.
Platform: desktop, Mobile Web, App
Users are served direct widget recommendations depending on the segment they are in and what they have seen so far.
Platform: Application
Amazon encourages users to mimic the buying behavior of customers who have bought an item that the user is currently viewing .
Platform: desktop, Mobile Web, App
Logged-in users have a rich set of recommendations with different strategies per user and per channel.
In addition, logged out users see a location booking symbol for where recently viewed items would be located, with a call to action to sign up and log in to see personalized recommendations.
Strategies developed in this section include
Platform: desktop, Mobile Web, App
On the cart page, users can access their stored items.
In this section, users are encouraged to click on “Move to Cart” if an item is in stock, and are served with “Prime”messaging, if available, and CTA for “comparison with similar data”.
Platform: desktop, Mobile Web, App
On the shopping cart page, Amazon presents an option to “Compare with similar items” below the product descriptions.
Users can click on this CTA and see similar products with an item in their shopping cart in case they want to edit or add another item to their shopping cart before checking out.
On mobile, the “Similar Items” widget is a dimmer that appears at the bottom of the screen.
On the desktop, an overlay widget appears when the user clicks on the CTA.
Platform: desktop, Mobile Web, App
The search results pages are modified for different users making the same search query.
The products are sorted differently, but the number of products displayed per page is the same.
Platform: desktop, Mobile Web, App
When a user arrives at the app/eshop and starts typing a query, a drop-down menu appears, consisting of product recommendations (simply the names of products) based on how many letters the user has typed.
Platform: Email
Amazon uses several recommendation strategies when sending purchase confirmation emails.
These strategies shall include recommendations based on
Each email has an average of 2 recommendations, and these are developed exclusively in sending emails.
Lartigiano increases revenue and cart using Artificial Intelligence!
See how he did it here!
Contact us now to see how the mobiplus shopping recommendation platform can give you the above features and increase your revenue. 30%.
mobiplus member Elevate Greece
Confirmed Innovation
Test out the Recommendation Engine for up to 20.000 euros in revenue from recommendations. No credit card required. Just enter your company information and we’ll contact you with all the details.