I recently watched a fascinating video by Jeff Su that completely changed how I approach data analysis. As someone who works with data regularly but lacks formal training in analytics, I was immediately drawn to his simple yet powerful framework that transforms ChatGPT into a personal data analyst—no technical skills required.
What struck me most was how universal this challenge is. Nearly all of us work with data regardless of our role, yet very few receive structured training in how to analyze it effectively. This gap between expectation and preparation creates unnecessary stress and inefficiency in our work lives.
Jeff’s three-step framework, called DIG (Description, Introspection, and Goal-setting), offers a practical solution that I’ve already started implementing in my own work. The beauty of this approach is that it leverages AI to do the heavy analytical lifting while we focus on asking the right questions and interpreting the results.
Understanding the DIG Framework
The DIG framework transforms how we interact with unfamiliar datasets. Instead of staring blankly at spreadsheets or wasting hours trying to make sense of raw data, we can use specific prompts to guide ChatGPT through a structured analysis process.
Here’s how each step works:
- Description: Have ChatGPT explain what’s in your data file as quickly and effectively as possible
- Introspection: Ask ChatGPT to brainstorm questions it could answer with your data
- Goal-setting: Direct ChatGPT to focus on specific objectives relevant to your needs
What makes this approach so powerful is how it builds understanding incrementally. With each prompt, your comprehension of the dataset increases until you’ve uncovered insights that might have taken hours to find manually—if you found them at all.
Putting the Framework Into Action
The Description phase is where we establish a baseline understanding of our data. I love Jeff’s example of starting with simple prompts like “list all columns in the attached spreadsheet and show me a sample of data from each column.” This gives us an immediate overview without overwhelming us with information.
Following up with requests for additional random samples and data quality checks helps identify potential issues before they derail our analysis. This proactive approach to data validation saves countless hours of rework later in the process.
The Introspection phase is where things get interesting. By asking ChatGPT to generate questions we could answer with the data, we often discover analytical angles we hadn’t considered. I’ve found this particularly valuable when working with complex datasets where the potential insights aren’t immediately obvious.
For the introspection step, ask “What questions do you think someone would want to ask about this data but we can’t answer due to missing information?” This surfaces gaps in our dataset and helps manage expectations about what insights we can uncover.
The Goal-setting phase ensures our analysis remains focused and relevant. I’ve experienced firsthand the frustration of spending hours on analysis only to discover I was answering the wrong question. By clearly articulating our objectives upfront, we can direct ChatGPT to prioritize what matters most.
Beyond the Basics: Making the Framework Work for You
While the core DIG framework provides an excellent foundation, I’ve found that adding a few personalized touches can enhance its effectiveness:
- Always ask ChatGPT to explain its reasoning when providing insights
- Request visualizations or charts to make complex patterns more accessible
- Challenge ChatGPT to identify potential biases or limitations in its analysis
One of my favorite strategies is to ask, “What are the key questions someone reading my analysis would ask and how should we proactively address them?” This anticipates potential challenges and strengthens my presentations before they face scrutiny.
Why This Matters for Non-Data Analysts
The ability to quickly analyze and extract insights from data is increasingly essential across all business functions. Whether you’re in marketing, operations, finance, or leadership, making data-informed decisions gives you a competitive edge.
What I appreciate most about Jeff’s approach is how it democratizes data analysis. By leveraging AI through a structured framework, we can all perform sophisticated analyses without specialized training or technical skills. This levels the playing field and empowers everyone to contribute meaningful insights.
I’ve already seen how this framework can transform the way teams approach data challenges. Rather than waiting for data analysts or making decisions based on gut feeling, we can now confidently explore datasets and extract actionable insights on our own.
The next time you’re faced with a spreadsheet and that familiar feeling of dread, remember the DIG framework. With the right prompts and a bit of practice, you’ll be amazed at how quickly you can transform from data novice to confident analyst—with ChatGPT doing most of the heavy lifting.
