Hi, my name is Mrudhulaa. I have about 3 years of experience in IT and full-stack web development. I’ve worked with clients on software design, development, and web application integration in both B2B and B2C platforms. I have some experience in retail and e-commerce, as well as customer analytics, user experience, web accessibility, and functionality. I had little understanding of how data is used to make decisions before enrolling in this course. The primary reasons for enrolling in this data science course are the upcoming demand for data science and my desire to learn how minor changes to the website can be leveraged and used to achieve favorable results for the business.
My job as a web developer entails creating web pages based on business requirements. I was assigned an ad hoc task to analyze the data collected from GTM (Google tag manager). I had to use GTM for almost all of the features on all of the pages. One of the benefits of GTM is that it allows us to easily track various aspects of our website, which allows us to better understand our audience and determine which areas of the site are performing well and which can be improved.
I used data layers in a variety of places on the page, including the login button, navigation links, cart button, add to cart CTA’s, zip code link, and so on, and this data is collected as a value to the variable set up by me via javascript and stored in the Google tag manager form. We can run this alongside our website to see if all of the tags are firing correctly on specific clicks. We can publish this GTM setup once all of these items have been checked. The final step is to design custom dashboards and set goals for the points being tracked. Aside from that, the collected data can be saved as a CSV file and analyzed further.
To analyze our site’s behavior, we used a K-means clustering model. We intended to capture insights like which product the people are more inclined to, how long they spend browsing our site, how quickly they leave a page after visiting, what features they check out, and so on, regardless of their geographical locations. As a result of the clusters, we made a few changes to our website for specific locations where the audience count is much lower, and these changes are much perceptible to all of our customers individually, for example, the background of the page, feature styles, price level changes, and setting up certain offers and discounts to attract a larger audience. This analysis and its findings have greatly aided businesses in expanding their customer base. We can perform numerous analyses on these, but they are limited to specific business lines.
As a web developer, I had the opportunity to work on this interesting project. Machine learning has undoubtedly disrupted various aspects of e-commerce and marketing. Marketers can now analyze large amounts of data and use it to retain and improve their overall customer base, as a result of this. Eventually, it impacted a 12 to 15% improvement in core web vitals metrics and a 20 to 30% surge in new customers. Also, we received a 5 on 5 rating from a recent survey conducted.
