As you advance in your career, you’ll encounter more content to read—it becomes the primary mode of communication in most organizations. I’ve found it useful to think about the material you’re reading and align it with an appropriate strategy.
Let me break down four distinct types of reading and how to approach each one effectively.
Type 1: Reading to Extract
The goal here is to scan quickly. This is where LLM tools like ChatGPT can be useful.
Common examples include:
- Reading through bulletins or reports from your organization
- Scanning newsletters or newspapers
- Reviewing meeting invitations for planning sessions
When you’re extracting information, you already know what you’re looking for—you just need to find the specifics. For instance, with a meeting invitation, you know you’re looking for a date and time. When viewing a map, you’re searching for a location. Train your eyes to scan for that information and ignore everything else.
Type 2: Reading to Understand
This is where people tend to get tripped up. You might encounter this when:
- Starting a new project or initiative
- Attending your first meeting on a topic
- Reading an article online
The key is to grasp the main idea as quickly as possible. The mistake I see people making? Starting from word one and reading straight through to the end. This approach is unlikely to be effective.
A More Effective Approach
Instead of reading linearly, look for strong information anchors:
- Diagrams or visual summaries if available
- Tables of contents that highlight key sections
- Visual representations like mind maps or flowcharts
Identify the top five points as quickly as possible and create your own mental model. For me, visualizing something like a flowchart works best—understanding the end output and the steps to get there. You can reach this level of understanding in just a few minutes without having to wade through a thousand-page document.
Pro tip with ChatGPT: Instead of just asking for a summary, ask it to give you the 10 or 20 most important concepts, then piece them together yourself. This requires more effort than requesting a summary, but you’ll end up with more useful working knowledge. Why? Because you know why you’re reading that article—you understand its relevance in a way ChatGPT doesn’t. Every AI summary reflects some point of view, but which point of view? ChatGPT will give you a middle-of-the-road answer. If you extract the top concepts and connect them yourself, you’re more likely to find something useful for your needs.
Type 3: Reading to Learn
This is different—you’re dealing with substantial material where you need to master the content. Examples include:
- Textbooks for a certification exam
- University coursework
- Preparation for professional certifications
A lot has been written about effective learning (I’d recommend the book Make It Stick), but here are some key principles:
Go heavy on the testing effect. If you know how you’ll be tested, keep testing yourself repeatedly. For example, when I did my UK driving theory test—a multiple-choice exam—I found an app with the kinds of questions I’d see on the exam. I did those questions over and over again. It might sound repetitive, but it changes how your mind approaches the material.
Match your practice to the assessment format:
- If you’re facing an essay exam, write as many practice essays as you can
- If you’re learning a new chapter and don’t know what will be tested, read through it once, then immediately write an essay about what you’ve learned
- Continue this pattern consistently
Using LLMs for Learning
The best way to leverage tools like Claude or ChatGPT here is to ask them to quiz you. They can create interactive quizzes on the spot that are customized. Claude has “artifacts,” Gemini has “canvas”—all these tools support this kind of interface. Even if they don’t have a special feature, ask for questions and then request scoring after you’re done. That’s how you learn.
Type 4: Reading to Apply
This comes into play when you’re learning something practical—for example, how to build a product roadmap for something you’re actively working on.
The single best thing you can do: get feedback as soon as possible. When it comes to application, nothing beats actually doing the work in real-time and evaluating its usefulness.
Key Principles for Application
Simplicity is your friend. If a tool is taking too long to understand, either:
- You’re not ready for that tool yet, or
- You don’t have enough context
Make it relevant. Take the simplest project that will have real impact and do it. For instance, if you’re learning to code, choose the simplest project with tangible real-world impact and work on that. Everything you learn becomes relevant to your project.
That’s my advice on the various ways of approaching different types of reading. Match your strategy to your purpose, and you’ll learn far more efficiently.