When we're looking to improve ourselves or society, a very common problem is that we're trying to do too much. We notice that there are many things we in principle could improve: our sleep, our training routine, our cognitive habits, our looks, our foreign language skills, etc, etc. More often than not, we take a far too indiscriminate and over-optimistic approach to such self-improvement work.
We fail to prioritise appropriately between potential improvements, partly because we don't consider opportunity costs to the extent that we should, and probably partly because some our shortcomings seem so important to improve on that we can't stand de-prioritising them.
We're over-optimistic regarding how many of them we can implement, partly because we underestimate their costs (we often fail to see that doing things in a new and unusual way is much more costly than the default approach), and partly because we're over-optimistic about how much energy, initiative, and attention we have, and think that we're able to sustain more comprehensive programs of self-improvement than is actually the case.
The same is true at the societal level. For instance, people notice that there are very many things that children ideally should know. By default, whenever they notice some such thing, they suggest it should be taught in school. But that's not a good heuristic, because there are endless numbers of things that children ideally should know, and only so much time. The same holds true in many other domains on the societal level. There, too, we fail to prioritise, and are over-optimistic about how much we can improve.
It's sometimes said that the job of the minister of finance is to say no. All the other ministers are looking for more funding to their department, but the minister of finance's job is to make sure that funds are used rationally, and that the budget is (more or less) balanced. That entails saying no a lot, and refusing to fund projects which, while not totally useless, just aren't worth it.
When trying to improve ourselves and our society, it can be good to think a bit more like a cautious minister of finance, who knows that we need to prioritise wisely, that our resources are more limited than many think, and that we therefore need to say no a lot.
Another lesson (from the over-optimism point specifically) is that often we should by default do things in the conventional way, unless there's a compelling reason not to. That's because it's usually more effortful to do things in another way than we think, and drains our (limited) attention. We need to save that attention to the improvements that really make a difference. Note that this argument is different from the "spend your weirdness points wisely"-argument.
That argument says that we should do conventional things for reputational reasons. Here I rather point to more direct costs of new and unconventional approaches.