When the data and anecdotes disagree, the anecdotes are usually right, which means you should examine the data to ensure you're measuring the right thing.

I emphasize the importance of qualitative feedback over quantitative data, as customer insights can often be more meaningful than mere numbers.

The Titanic had big data suggesting no icebergs were present, but anecdotal evidence of an iceberg was crucial and ultimately more reliable.

Talking to customers provides more depth and information than analyzing data points in a graph, which is critical for building intuition about your product.

Proprietary data can be less valuable than the insights generated by infinitely smart agents using public data, as the latter can synthesize new information more effectively.

When making decisions, sometimes you have to trust your gut over the data, especially when you understand the end user's needs.

Books are often more valuable than bite-sized information from blogs or social media, as they contain wisdom that has stood the test of time.

I think the analytics in sports are table stakes; the real value comes from how you use the data and the questions you ask.

It's important to observe how people use software in real time, as this can provide valuable insights that metrics alone cannot capture. Relying solely on metrics to understand user behavior is like observing raw data from a radio telescope instead of experiencing the night sky directly.