In the world of e-commerce and e-services Customer Experience personalization is thoroughly dependent on data. After you have acquired the right data you can shape the close-to-perfect overall customer experience by harnessing machine learning and data science.
Let’s have a quick walkthrough of some examples of customer personalization by visiting Amazon and Netflix.
1. Product recommendation
Product Recommendations are essential on a well-personalized e-commerce page, should it be product- , category- , home- or other kind of page: the presence of item-to-item collaborative filtering-based recommendations is a basic requirement, but, after the recommendation engine learnt the customer’s preferences it will display more and more personalized offers.
Product recommendation is an enormous topic, we are going to detail it in this dedicated unit.
2. Product listing personalization
Prior to this product listing our trial user visited several product pages displaying home appliances.After that, he made Amazon list Today’s Deals - in the product listing the home appliances appeared overrepresented among the first row products.
3. Search personalization
I had visited a few ‘Felt Furniture Pads’ - product pages, then I started searching on products starting with ‘F’ : The Amazon search field remembers and displays the search keywords referring to my preferred products.
4. On-site personalization
Displaying the user’s name on highlighted places are typical examples of on-site personalization.
5. Social personalization
Retargeted Amazon Facebook ads displaying the user’s preferred brands.
6. Email personalization
Recommendations sent in email. The products listed here are in accordance with the user’s previous purchases.
7. Available recommendations - Offering room for customization
However Customization doesn’t belong to the category of Personalization, it’s important to let the user adjust (customize) what kind of products he wants to see beyond those that are offered to him by his tracked events and analyzed preferences.
You can often encounter irrelevant experiences like being recommended an already-bought product by an e-store, or being given recommendations to buy cosmetics you have never displayed any interest in. You can also stumble upon retargeted listings about products you viewed at some point but finally, you turned them down.
This problem called irrelevance can be very severe when there are plenty of products to offer. Using the powerful subset of personalization (Product Recommendations by user preferences) Netflix solved this problem among the first players of the industry.
At least 80 % of the Netflix TV shows are discovered through the content recommender system that is known as one of the most important component of website personalization.
After watching some movies in Netflix, the site will recommend you another ones on the shelves with the following lines: “Because you watched show A, you will like B,C and D.” - This content is driven there by the black box having the recommender engines inside: recommender engines use machine learning and algorithms to gain insight into the viewers’ preconceived notions and display them the shows that they will be interested in.