If you have 2+ groups with opposing views, each 3.5%+ it's pretty clear that at least one of the 3.5%+ groups will fail.

Others here note it's really "3.5% if there's no one seriously opposing their objectives" but in my opinion that's a meaningless rule. Of course in those cases non-conflict resolves the issue.

This is far from meaningless, because if you are too far below those 3.5%, you'll fail to make a change for the better, despite having a good cause with no real opposition.

Those 3.5% are encouraging for all social movements, who suffer (and/or have friends/family who suffer) from some issue in the system, have perhaps developed a good plan out of it, but think they are too small to make a difference.

Success doesn't have to mean getting your way, but rather making a meaningful change in your direction. Even opposing groups often can find a way so that both get a better situation. For instance, taxes can overall be lowered while teacher salaries can increase on average at the same time, if excess money is taken from other activities.

Yeah but that probably isn't going to what the original research is saying. Society is basically run by a tiny fraction of people (1-5% of the population range) and the rest are just along for the ride. Democracy is a major innovation where the majority has to nod along every few years or there is a mix up in who in the upper class gets to sit at the top of the tree.

From that perspective it becomes clearer what a 3.5% rule is getting at - 3.5% of the population mobilised is enough to overwhelm any ruling class that isn't on top of its game, especially if mass shooting of people is still of the table or if the 3.5% includes a lot of people from the upper classes. It isn't about whether an issue is supported by 3.5% of the population or more, it is a question of whether that fraction of society is actively trying to topple a government system.