Posts

Cohort tables offer valuable insights into customer behavior by grouping them based on shared attributes, in this case, acquisition date or date of first purchase. They are fantastic for identifying whether you have a retention problem or a user acquisition problem and can allow marketers to understand user behavior through experiments. Learning to read these tables can be tricky at first but effectively empowers you to analyze customer engagement, retention, and acquisition trends. ...

King County, WA Housing Insights Line Chart: The line chart displays daily prices from May 1, 2014, to May 1, 2015, illustrating a general upward trend over the year with some fluctuations. Housing prices rise from an average of $0.5M to $1M by November 2014, peaking at $1.3M in May 2015. The zip codes with the highest average price, ranging from $0.9M to $1M, include 98109, 98105, 98004, and 98040. These zip codes, situated close to Lake Union, the only water body within city limits, are still connected to the mainland, possibly making them desirable locations for homes. This observation is supported by the fact that Vashon Island, despite being surrounded by water, is one of the cheapest places to own a home, likely due to its isolation from the mainland and possibly poorer infrastructure. ...

As a Data Scientist at the ShapeUp Project, I’ve been working on data analysis projects at Shape Up The Vote Data Project that activate barbershops and salons as voter engagement hubs in 13 states with 5,500 participating shops by 2024. We use relational organizing principles and remote communication through text and phone banking. My role involves collecting, cleaning, and enriching data on potential participants to identify the best shops to contact. ...

When writing functions in Python, you may sometimes find yourself assigning a function parameter to a new variable. While this can improve readability and functionality, it’s not always necessary. In this article, we’ll explore when and why mapping parameters to new variables makes sense—and when it doesn’t. 1. Preserving the Original Parameter (Avoiding Side Effects) Some data types in Python, like lists and dictionaries, are mutable, meaning that modifications inside a function can affect the original object. If you need to modify the data but want to avoid altering the original input, creating a new variable is a good practice. ...

Choosing between long and wide data formats can make or break your workflow—especially when it comes to analysis, visualization, or preparing reports. This guide will help you figure out the best structure for your dataset based on your goals and tools. 🧠 Start With the Right Questions Before reshaping your data, ask yourself: 1. What are you trying to do with the data? Analyze it (e.g., run time-series or trend models)? Visualize it (e.g., bar charts, line graphs)? Use it in tools like R, Python, Excel, or Tableau? 🔧 Pro tip: Tools like ggplot2 in R or pandas in Python often prefer long format. ...