Google Sheet Hacks, APIs, Python , and more

Google Analytics recently announced the release of GA4. I have not tried it yet or move any GA instance to the new version yet but it looks fantastic. I have launched web apps on Firebase and it does look very similar on the surface to the analytics platform from there. I think the biggest news is that you can (finally) export your data to BigQuery (https://support.google.com/analytics/answer/9358801?hl=en) which was exclusive to paid versions of GA previously (I believe the you will still have to pay for BigQuery charges but please do your own research on this I can be totally wrong).


One of the goals I have for 2020 is to bring at least one of the machine learning model I have on my laptop into production on the web.

My initial thought was “how hard can it be?”.

  • Convert your Jupyter Notebook into a Python file
  • Maybe refactor some of the spaghetti code into functions to look more professional
  • Pickle your ML model
  • Spin up a simple Flask App, create a REST endpoint for your model, push to Github and deploy to Heroku — viola!

While this is a completely legitimate way to just get something onto the web, my…


80% of your work in data analysis and machine learning is wrangling and cleaning data which can be very tedious work. Automation of data extraction, data enrichment, and data cleaning with APIs can save you a lot of work doing manual extraction.

The focus in this article will be on how to interact with APIs in Python and we will be working in the context of the Singapore public housing data set and explore the Bid-Rent Theory in the housing market. …


Important note: This was created as part of my own personal learning process for data science in python. I find it extremely helpful when i write this down to help me learn better and faster. This was part of the course on DataCamp and the code is based on the course and other online resources I used. Please visit DataCamp for the original syllabus. I am not affiliated with DataCamp in any form and only use their resources as a learner.

On the final part of our customer segmentation journey we will be applying K-Means clustering method to segment our…


The last time we analyzed our online shopper date set using the cohort analysis method. We discovered some interesting observations around our cohort data set. While cohort analysis provides us with customer behavior overtime and understand retention rates, we also want to be able to segment our data by their behavior as well. Today, we will be exploring the popular RFM model used by retails such as Sephora, blending in-store and online purchases to segment their customers for better personalized ad content. I would highly recommend following the Data Science at Sephora blog for more in-depth data insights. …


Important note: This was created as part of my own personal learning process for data science in python. I find it extremely helpful when i write this down to help me learn better and faster. This was part of the course on DataCamp and the code is based on the course and other online resources I used. Please visit DataCamp for the original syllabus. I am currently no affiliated with DataCamp in any form and only use their resources as a learner.

It is important for a business to understand if their products are selling well. It is arguably more…


Building Trust in your Marketing Data and Analytics — Performance, Accountability, Transparency

I was listening to a great podcast recently by Chris Butler on how to build trustworthy AI products. I would recommend this to anyone working in Product Management to have a listen. This got me thinking — the same is true for marketing data and analytics today. Building good data models and building trust of the people that will be using it must go hand-in-hand.

The Problem with “AI” in Marketing

Oftentimes poorly implemented models and the lack of explanation of these how they are derived not only…

Yexi Yuan

Business Intelligence Analyst for Direct-to-Consumer products.

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