I've learned that patience and testing have to work together. You don't want to kill an ad after a tiny amount of data, but you also don't want to keep spending money indefinitely on something that's clearly underperforming. Finding that balance took me quite a while.
 
Back
Top