What defines "good governance?"

In further discussion about whether constitutions were good, necessary, the devil's work, etc., a somewhat strange issue popped up: no one seemed to agree upon what makes for good governance. There seemed to be no convenient way to compare countries. When the best categories you can come up with are "Pretty Successful, Hasn't Exploded Yet" and "Hell hole" you really have hit a road block in the debate.

Thinking upon this further, there does seem to be a bit of a gap in even agreeing upon what government's prime metrics should be, much less how to measure them.
Should government try to maximize freedom? Even if you can agree upon that (and strangely that is not always possible) defining freedom is a little difficult. Measuring whether US or UK is more free is far more so.
Is the mark of a good government a strong economy? That should be easier to measure, but as it turns out isn't too obvious either. You can use per capita GDP, but then can you say a poor country that is growing quickly has a better or worse government than a relatively rich country that grows slowly? Maybe the poorer nation is fantastically governed, and just needs time to catch up. Throw in the fact that GDP isn't even a good measure of wealth, and how easily economic numbers can be cooked by less honest regimes and that is no longer a good estimate.
People sometimes cite crime statistics, or health care, or some other factors, but those suffer many of the same problems as GDP. Crime rates might be useful on a city by city, county by county level, but at the national level you run into problems of categorization, differences in demographics, and even just straight cultural differences that cloud the issue. Health care is no less awkward. The UK and most other nations measure life expectancy differently than the US, due mostly to how they treat death in the first few days of life, and that makes direct comparison remarkably difficult if you don't know all the subtleties and correct for them.

In the end, the best way to compare that I can think of is to look at net immigration between two countries. If more people are moving from Country A to Country B than from B to A, you can probably assume B is considered to be a little better across all possible criterion. This also has the advantage of being a pretty clear number, with data over time. One can argue that this is because of B having a better economy, less civil strife, or just looser immigration laws, but everything but immigration laws can perhaps then be bundled into "outcomes people wants from their government." (In fact, maybe looser immigration laws too.) If people want a multitude of different things from the government of the territory they choose to live in, perhaps it is best just to look at what bundles of those things they choose, and then consider why some bundles are chosen over others.