⚠️ This column shares the prompt creation process for free educational purposes and does not constitute investment, legal, or financial advice.** AI execution results may contain errors, so independent verification is essential. All responsibility for their use lies solely with the user. It cannot be directly used for business decisions. [See all essential checklists below the 30-week timeline before use[]](https://www.notion.so/30-29f84ddcec3c80a1a87eda9883d81453?pvs=21)
In Week 3, we narrowed 25 sources down to 14 through a four-step filtering process. However, we were missing something crucial. Defining the information (content) was the priority, not finding the source (source). Since each source published a wide variety of information, I calculated that even if I only searched 10 items from each source, I would have to search 14 × 10 = 140 times.
140 web_searches would mean checking and summarizing approximately 1,400 articles. The problem was that there was no standard or method for mapping and organizing all 140 pieces of information. I reconsidered, "Why did I even bother collecting policy information?”
My goal was to predict the next tariff item to expand my investment portfolio. So, the question was wrong. "What information does each source release?" is a source-focused question. "What information is needed to predict the next tariff item?" is a goal-focused question. From then on, I changed my thinking and redefined my goals.
I decided to focus on my true goal of predicting tariff items, and I began organizing the information before even gathering it. While studying investment, I wanted to find the best investment strategy, and to achieve this, I followed the same approach of comparing each book I read to the strategies covered in previous books.
The key is simple: understanding the structure first. After reading nearly 300 investment books, I realized that investment strategies can be broadly divided into two categories: trading and investing. However, when broken down into specifics, no single successful investor's strategy was identical to another's. However, their strategies could all be explained by returns, entry points, win rates, take-profit margins, and stop-loss margins.
Based on this, I was able to measure the past returns of every investment strategy in the world, and the most important factor, of course, was return. While policy information is complex and government agencies are diverse, what ultimately matters to me is predicting the next tariff item. I organized my thoughts by clarifying that I was gathering information for my own goals.
In other words, rather than simply collecting policy information released by government agencies, I shifted my focus to proactively searching government agencies to gather additional information, such as tariff rates and tariff periods, based on tariff items.
First, I decided on my search targets. It's impossible to examine all tariff policies, so I focused on semiconductors, a representative industry, to establish a system. I believed that if I could learn how to collect policies using semiconductors as an example, I could apply the same approach to other industries. Since my goals had changed, I forgot all 14 existing sources and started anew.