Weirdly enough, the first thought that comes to mind is actually the Sun Tsu axiom about knowing yourself and your enemy.
If you’re building a world from scratch, you need to look at this one. You can see us do this on a fairly regular basis, especially when people are asking about putting together a fantasy setting.
Study history. Study the time period (or periods) your setting is roughly based on. Learn what you can, extrapolate what you want.
Further, study “why?” When you’re looking at a given era, pay special attention to the reasons behind the decisions people made. Find out why forces were arrayed the way they were. Find out why the weapons of the era were used, instead of others. Why some cities were valuable, and others weren’t, especially how those values changed over time. Look at why people’s value systems developed, and how they understood their world, right and wrong, ethics, morality, religion. Look for the causes that lead to their understanding (this one is a bit of a tall order, but, seriously, try it).
Studying “why” does two things: it lets you start to understand what will change when you start tweaking the world, and it helps you to understand people a little better, which is incredibly important as a writer.
Take a cause and effect approach to building your world, rather than the effect preceding the cause. A lot of the time, when we start building a world, that’s you, me, and probably most people reading this, it comes from a basic premise, “here’s a cool thing I want for my world.” On it’s own, that’s fine, it’s a starting point. But, immediately after that, you need to decide what caused that, and then, after that, resist the impulse to add another “cool thing” because you want it. Instead start looking for other consequences that come from your first cool thing, and expand out.
Because the great cool thing happened, people reacted (preferably like people), which lead to this other thing happening, which lead to people reacting, and more consequences, more reactions, more consequences.
You could call this hierarchical world building. You start with your basic cause and effect in mind, and then start working out all the thousand little cascading events and consequences.
Finally, people aren’t smarter now than they used to be. There’s a real truth to the line about standing on the shoulders of giants. When you’re building your world, don’t resort to people being idiots in order to explain their behavior. They may not have all the information, but they’re going to try to solve the problems they’re facing to the best of their abilities. If they’re ultimately working against their own interests, it’s going to be because of things they don’t know or understand, in their world, not because they’re stupid.