Muffaddal has 3.5 years of experience working with web analytics and user behavior data. ❄ https://www.linkedin.com/in/muffaddal-qutbuddin/❄

A detailed guide on how item-based recommender systems work and how to implement it in a real work environment.

Image for post

Recommendation systems have been around us for quite some time now. Youtube, Facebook, Amazon, and many others provide some sort of recommendations to their users. This not only helps them show relevant products to users but also allows them to stand out in front of competitors.


Monitor how long students spend time on Thinkfic lessons in Mixpanel.

Image for post
source pixabay, by anncapictures

Thinkfic is a platform to make it easy to create, market & sell online courses. Everything you need to run your education business under one roof.

However, one thing it lacks is the ability to track the session duration for each student. You can’t monitor how much time students spent on courses.

We can leverage Mixpanel analytics for this purpose. It allows analyzing user's (students in our case) behavior across your sites with a real-time cohort.
The best part is Thinkfic has direct integration to Mixpanel.

Here is how it will look in the Mixpanel


How to Send SendGrid's email engagement data to Google BigQuery

Image for post
Source Pixabay, by ribkhan

Capturing data from multiple sources is the key to creating a rich data warehouse. And it becomes even more crucial when it comes to online businesses as the data is scattered across a plethora of platforms used to run the business.

SendGrid is one such marketing platform that delivers transactional and marketing emails to help keep users informed and engaged.

We will go over different methods to store SendGrid email data to our data warehouse tool.

SendGrid Webhooks

SendGrid provides several ways to export its data out. One such way is to use SendGrid webhooks. Webhooks are automated messages sent from apps…


Automated data pipeline to move Cloud SQL data to BigQuery

Image for post
source pixabay, by Dimhou

Cloud SQL is a database service from Google Cloud Platform that makes it easy to set up, maintain, and manage relational databases.

However, It isn’t built for heavy analytics operations. You need a warehouse service such as BigQuery to do advanced analytics and machine learning tasks such as performing RFM analytics using BigQuery ML. For this, you need to move your cloud SQL data to BigQuery.

In this guide, we will build a data pipeline to send the cloud SQL data to BigQuery in an automated fashion.

Data Pipeline

For our data pipeline, we will make the use of two GCP components…


Recommender System to suggest similar items to users on product detail pages

Image for post
source pixabay, by bboellinger

An e-commerce company reached out to you to build a recommender system to suggest three items on their product detail page. Their analytics systems have around 30% of the user’s purchase history and only 20% of the products have users rating history. What method will you use for this type of problem? What will be your evaluation technique?RMSE, Precision, MRR, nDCG, or any other?

In this article, we will build a recommender system for the product detail page to recommend items to users based on their rating history.

We will evaluate Eight different models to get ourselves the best working…


Get similar products to recommend using recommenderlab in R

Image for post
source pixabay, by lumen

Finding similar products to suggest is a tough nut to crack, especially when we have a large number of products. We want to suggest similar products in our number of marketing and product activities, such as in promotional email marketing.

Here is the super art, I build using autodraw, illustrating a marketing email having suggestions for similar products with the main product.


How to create dynamic links in Firebase.

Image for post
Source Pixabay, By Skitterphoto

Firebase Dynamic Links help to control where a user lands depending on whether they have an app installed or not. Having such behaviour gives a richer user experience when interacting with marketing campaigns.

In this guide, we will go through different Firebase Dynamic Links components that you need to consider when creating a dynamic link. These help to ensure that users are directed to the desire landing page/app section and that marketing channels remain intact throughout the conversion.

Dynamic Link Components

There are four components of a dynamic link to keep in mind when creating a Firebase Dynamic Link:

  • What to do when…


Track user activities on WebView in Firebase Analytics.

Image for post

More and more mobile applications are utilizing power WebView to customize the user experience on runtime. And it is all the more important to track users’ activities on the WebView.

However, Firebase analytics’s SDK doesn’t support sending events from WebView pages of the mobile app. So to use Analytics in WebView, its events must be forwarded to native code before they can be sent to analytics.

So,
First, we have to pass the events from WebView to native code.
Second, fetch the passed events in the native environment and send it to Firebase Analytics.


Execute multiple queries one after another in sequence in BigQuery

Image for post
By Dakub, Souce Pixabay

Bigquery is a fantastic tool! It lets you do really powerful analytics works all using SQL like syntax.

But it lacks chaining the SQL queries. We cannot run one SQL right after the completion of another. There are many real-life applications where the output of one query depends upon for the execution of another. And we would want to run multiple queries to achieve the results.

Here is one scenario, suppose you are doing RFM analysis using BigQuery ML. Where first you have to calculate the RFM values for all the users then apply the k-means cluster to the result…


User Segmentation using RFM analysis in BigQuery ML and Visualization in the Data Studio.

Image for post
Image by Gerd Altmann from Pixabay

Suppose you want to remarket your users for conversions and reactivations by creating segments based on their website app activities using the google analytics data stored in BigQuery.

In this guide, I will show how to implement one of the popular techniques to segment users, and visualize the results to help make important marketing decisions.

The technique we will go through is RFM analysis. We will be using BigQuery ml to create segments and the data studio to visualize the results. Here is how it will look.

The dashboard of segmented users, by Muffaddal

In this guide we will:

First…

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store