Local maxima vs global maxima

In optimization, a local maximum is the highest point within a limited neighborhood, but not necessarily the highest point overall. A global maximum is the highest possible point across the entire problem space.

This concept is a useful analogy for business strategy. A company can become very good at optimizing a specific part of its business, reaching a local maximum. For example, a company might perfect a particular feature or marketing channel.

However, this can be a trap. By focusing too much on local optimization, a company might miss a much bigger opportunity—a global maximum. This could be a new product, a new market, or a completely different business model.

This is a key limitation of a purely data-driven approach. Data-driven optimization is very good at finding local maxima. It can tell you how to improve what you're already doing. But it can't tell you that you should be doing something completely different.

To find the global maximum, you need human creativity and strategic thinking. You need to be willing to step back, question your assumptions, and explore new possibilities. This is where a data-informed approach, which combines human insight with data, is more effective.