As part of an ongoing series, I’m trying to put theory into practice and help people reading this column apply AI to their daily work. In our first article, we explained how to get started using text-to-speech technology in a few easy steps. Today we’ll be looking at data analytics, an increasingly popular use case (though not without its drawbacks).
How does AI interact with spreadsheets?
You’re probably used to having a spreadsheet-savvy data analyst on your team. This analyst is your go-to person when you need to do custom Excel work. I was once such a person. Performed large-scale data analysis using a combination of tools such as Excel, MS Access (using SQL), and SAS. However, with the advent of AI, the process has become much simpler. Your best analysts will have a much greater impact because they no longer have to waste time on unnecessary tasks like data cleansing.
Here’s a rough idea of how AI works:
- Data analysis: AI deciphers the structure of your spreadsheet and identifies rows, columns, headers, and data types.
- Pattern recognition: AI uses machine learning to recognize patterns and correlations in data that are not immediately obvious to human analysts.
- Data cleaning: Automatically detect and correct errors and inconsistencies such as missing values and outliers.
- Generate insights: By applying statistical models, AI derives insights, predicts trends, and makes predictions based on data.
In other words, a good analyst who knows how to ask the right questions can generate more insights than an analyst without AI.
How can I try this?
- Create a Chat GPT Plus account. There are other options available here, but most people are likely familiar with Chat GPT, so use the one you’re familiar with.
- Search for the Explore GPT tab in the left pane. Once there, select Data Analyst GPT (pictured) and a chat interface with it will open.
- Select the dataset you want to work with and download it as a CSV (or any other format, really). If you’re looking for sample data to use, visit Kaggle for a great repository with lots of interesting stuff.
- Upload the CSV to your chat window and prompt Chat GPT to work with your spreadsheet. I often use a prompt that says, “Read the CSV you are about to upload, remove any errors, and prepare for a series of questions to ask for further analysis.”
- ask a question. Here’s the key – ask good questions. An example is, “From this data he reveals five non-obvious insights.” “Help me find long-term trends that I tend to miss.”
In the example below, we downloaded a dataset about New York real estate from Kaggle. We then uploaded it to Chat GPT and asked it to reveal some insights. Here are the results obtained:
Of course, the correlation analysis is not very deep, but I was able to gain more insight very quickly and without needing much additional prompting. One obvious drawback I’ve noticed, as a disclosure, is that as datasets grow, there can be difficulties with both the speed and accuracy of analysis.
How should you leverage data analytics today?
There are many practical use cases for data analytics today that you should consider using. Examples include:
- financial forecasting – Ingest existing financial information and see if there are trends, such as seasonality, that can be more easily identified with AI.
- better human resources practices – Reduce employee turnover, retention, and improve hiring processes by understanding what goes wrong.
- Competitive analysis – Track your competition by evaluating your rivals’ SEO strategies and more.
The options are endless, but you would be making a big mistake if you didn’t start using some of these tools sooner. They only get exponentially better from here.
follow me twitter Or LinkedIn.
