Often I will hear, "Well once we start targeting their wives and kids that will knock the snot right out of them!" which is both simplistic and wrong. You don't win wars by killing civilians, you win wars by making the enemy military inconsequential to the outcome (in an insurgency you make the military inconsequential by winning the hearts and minds of the populace). You can do that by destroying their Army, or by dropping Nukes on their industrial base, but simply killing civilians has never been a major factor in winning a war. The London Blitz and the firebombing of Dresden should give you an idea of how ineffective targeting civilians for lethal effects is in the history of war.
On the flip side, targeting civilians for non-lethal effects can produce amazing results. The peace movement in the US during the Vietnam war was the Communist second front. The attitudes of the British citizens during Ghandi's peaceful crusade is another good example.
Do you want to know how to make a fanatic? Someone who will devote their life utterly to the cause of destroying you? Take something irreplacable from them, like their wife or children. But it is a second/third order effect.
Cause 1: Wife and kids assassinated.
Effect 1: Jack Booted Thug very angry.
Cause 2: Jack Booted Thug very angry.
Effect 2: Jack Booted Thug works harder to punish those he feels responsible.
Cause 3: Jack Booted thug works harder to punish those he feels responsible.
Effect 3: Less freedom of movement and maneuver for insurgent forces.
When you are trying to figure out second and third order effects, the primary effect becomes the secondary cause, which causes a secondary effect, which becomes a third order cause. You could follow this down the rabbit hole as deeply as you want to go, but generally the third order is as far as anyone wants to look into the foggy crystal ball.
On the flip side, there is the law of "unintended consequences" which basically states that every time you plan to get an effect that is advantageous for you, you run an equally high risk of creating another effect that is negative to your cause. And because you aren't smart enough to predict what that is, it will blindside you because you didn't have the information in you needed to avoid that consequence.
Imagine the negative consequence if that wife of the Jack Booted Thug was a well loved volunteer at the local Veteran's Hospital who took her kids in to play checkers with those in recovery? Unintended consequence becomes that the insurgents are viewed as murdering scum and the JBT is viewed as a legitimate victim who has a reasonable passion to bring justice to the villains who wounded his heart.
Now imagine if you inserted pro Freedom arguments in the head of the wife, and pro freedom arguments in the heads of the children? Can you measure that effect as the JBT has to argue his position at the dinner table that he really isn't a bad man? No, you cannot measure that effectiveness in terms of enemy killed or material destroyed.
But winning the war isn't about destruction, it is about making the enemy irrelevant to the outcome. You can do that by destroying the Army, but that rarely happens in an insurgency. We never defeated the British Army in our Revolution, we simply defeated the small portion that was stationed on the Continent. The British eventually took their ball and went home until the War of 1812.
You can apply the 2nd and 3rd order effect methodology to any tactical mission, and sometimes you can see strategic consequences arise (think Abu Ghraib or Mai Lai). Sometimes they aren't so clear cut and you have to give your best guess into the murky reactions of the enemy to try to optimize outcomes in your favor.
I have seen the consequences of military units that focused solely on lethal battlefield effects. They create more insurgents than they kill, and they manage to kill a lot of them (or so they think, insurgent corpses look an aweful lot like civilian corpses). The units that focus heavily on non-lethal effects, and use lethal effects to enhance security where only absolutely needed have been much more successful.