Algo-Based Website Personalization

Algo- Based website personalization is a subset of website personalization, in which instead of the product offerings being personalized on the basis of rules set up by the marketer, the personlization is provided out of the box using machine intelligence and AI/ML capabilities.

Under the Hood: How it works?

The first question that pops up in someone's mind is how does the algorithm work?

So the algorithms is loosely based on YouTube’s recommendation algorithm

It is a neural network with the base layer of collaborative filtering. The collaborative filtering algorithm essentially tries to match a customer’s interaction pattern to an outcome- people who followed this pattern in the past (clicked on products 65,18, 32, purchased products 456 and 43, and added product 365 to the cart) bought product 86. This current customer has a similar pattern, and hence she is likely to buy product 86

The next layer adds context – people in Istanbul in June bought more of these products and less of these. The given person is in Istanbul and it is June

The third layer adds temporal information. Essentially, it starts giving more weightage to the more recent interactions of the customer, and learns what the ideal decay rate should be
The last layer adds repeatability to the algo – how many times should you show a particular product to the customer before giving up.

Functionalities Provided

Page Reordering:

Product listing page: Personalized rows/page dependent on category/sub-category/sub-sub category

Deal pages: Personalize your sale pages based on which products the customer is likely to purchase from the sale

Personal Boutique :

Curated personalized product listing page based on your previous interactions

Liked/dislike feature: Interactive session featured on curated personal boutique or any Boxx widget where you can like & dislike the product.

Case 1: if you dislike, that particular product won't be recommended to you at any point

in widget

Case 2:If you like, that particular product would be added to Liked Product Page.

Liked product page: Appended list of products which you have liked during Like/Dislike feature interaction

Widgets:

Homepage: Most popular products widgets
Product display page: FIltered personalised recommended product widget based on your previous interactions

Exit Intent:

Pop-up: An exit intent widget, once you exit a designated area of a screen, a popup would be seen on the screen with personalised product recommendations.

Use Cases

  • Category page re-ordering: Products are re-ordered on the listing page for EACH customer based on her likelihood to click. For any customer, the products that she is likely to buy more are shown on top, while others are shown further down. Leads to more people visiting PDP page, which in-turn leads to higher conversions.
    To enable mixing of our algos, and your merchandising team’s logic, we usually mix the product listing - we won row numbers 1, 3 and 5, while your teams own the rest of the rows. We are flexible in this, and would discuss with your teams to manage

  • New conversion-optimized boutique pages: New SEO friendly pages are added to your site, with products personalized to each customer. These pages are accessible through the main menu, or through “view more” buttons on the widgets. These pages provide optional feedback buttons that are used to gather real-time feedback from customers and provide them with alternatives in real-time. Leads to a superb customer experience, more people visiting PDP page, which in-turn leads to higher conversions. Also boosts organic traffic.

  • Recommendation widgets: These are context-aware widgets that show personalized products to each individual on home page, PDP pages and Cart page. Leads to higher conversion rates.

  • Other widgets: Other optimization widgets are strategically placed within your website to compliment the recommendation widgets. These include “New arrivals”, “Bestsellers”, “Recently viewed” etc. Leads to higher conversion success of recommendation widgets.

  • Onsite popups: Popups trigger automatically whenever a particular visitor feels lost or is about to leave. This guides her in the right direction, thereby helping decrease the bounce rate.

Listing Page on CEE Panel

The widget numbers are now also shown on the CEE(please refer the image blow). You can navigate to this page from Campaigns -> On-Site Engagement -> Onsite Recommendation

The first level metrics tracked in the main report on the Onsite Recommendation page are as follows:

  • Traffic
  • Product Views
  • Add to Cart
  • Purchase
  • Revenue

Along with Revenue Attribution of your total revenue, described as "out of" in the tabs

These are then detailed out with metrics of each of your pages, Home page, PLP, PDP, etc

The chart displays all the active widgets on your page and plots them against the metric you choose:

You can choose the widgets that you want plotted along with seeing them at a Day, Week and month level

The metrics as follows:

  • Product seen
  • Product Click Rate
  • Add to Cart
  • Purchase
  • Product Revenue
  • Product Clicks
  • Add to Cart Rate
  • Purchase Rate
Details PageDetails Page

Details Page

Widget comparison

You can also go ahead and compare widgets of a particular page amongst them selves using the compare **widgets **option

Compare them using any of the metrics of your choice from

  • Product seen
  • Product Click Rate
  • Add to Cart
  • Purchase
  • Product Revenue
  • Product Clicks
  • Add to Cart Rate
  • Purchase Rate

Adding Comparison are as simply clicking on "Add new" and choosing your comparison metrics. Shown on the following image:

📘

Data Retention

Please note your data is retained for a max period of 4 months and are available on the dashboards and downloads as such.


Did this page help you?