I've been playing GalCiv II for a few weeks now gradually increasing the difficulty level. One cheese tactic I have found is going straight for engine and sensor techs and building fast surveyors on cargo hulls. With three or four of those I can get the majority of anomalies on a huge map with three opponents. I'm easily winning my first game on challenging using that tactic, although I did have the number of anomalies set to common. I must have gotten close to 20,000 credits that way, which totally smoked the AI which could expand as fast as me, but couldn't fund research while expanding. In the future, I will set anomalies to the lowest setting.

I always do the following:

Play a custom race, since choosing one just limits your potential number of opponents. I put points into morale, research population growth and research. Things like luck and trade don't seem to be all that useful.

I've been playing huge maps with three opponents, since more just means some will be very weak compared to others.

Set inhabitable planets and stars to common so that nobody is starved for planets to build on.

Turn off tech trading. I just don't like it since most cheese straegies revolve around tech trading.

I've started setting tech rate to fast since it is easy to get to the end game with half the tech tree untouched on normal.

Expand as fast as possible (no brainer), building faster colony ships as I go along until there isn't much left to take.

For a build que I always use some variation of:
Factory, factory
Lab
Market
Starport
Lab
Market

Then vary this according to the situation. If I'm doing well financially and in the tech race, I'll concentrate more on factories since those are almost always useful. I dislike cheese like building a factory on every square on some planets while filling another one with markets, with the various capitals being the exception.

I have learned to aim for a max population of about 18b, since more just become unhappy. Is there some great advantage to having 30b that I'm unaware of? I know you get more income, but is it worth the happiness hit? Most planets with mid to high population will need at least one entertainment improvement.

I will give in to cheap demands from aggressive opponents but refuse expensive ones.

I concentrate on constructors in the early mid game and economic starbases. I try to set them so that each touches more than one planet. Just as a no cheese rule, I don't dogpile starbases, limiting each planet to at most two economic starbases.

I try to hold off building much of a military until I can get medium sized hulls, since the smaller ones die very easily. When I get large hulls I almost never build medium hulls any more. I never use the stock ships. They stink big time. OTOH, as a balance issue I have found that a fleet of four large ships with good offense and defense tailored to what the enemy is using to be almost too easy. No AI can outthink a player, so wars are usually one-sided.

What would be nice is opponents who are nearly equal without being overpowering. One thing I have noticed about all 4X games is that the outcome is more or less decided very early. If you are competitive in the early game it is hard to lose the end game. If you are dwarfed in the early game, you face a titanic struggle from there on. The problem with changing difficulties is that it is easy to go from one extreme to the other. I'm not sure which is worse, winning too easily or being ground to dust by an opponent that is twice as large. I wish there were options for somehow tailoring the difficulty level differently in early and mid game.