With access to ever more precise PPC optimization tools, it can be tough to resist tweaking every minute detail. Like a child let loose in a candy store, refraining from touching and trying everything can take serious willpower. Most of the available tweaks are great ideas individually, but as a whole they can possibly impair performance.
Typically, data needs to accumulate in order to make decisions. On a large PPC budget, a lot of data can be gained quickly, however making daily or even weekly tweaks can derail an account's success when over-optimization occurs. Google has actually suggested 2-4 weeks worth of data be collected before considering data optimizing on a particular area of an account. While 3 or even 7 days might seem like enough time, many times this is still too small a window to actually have any actionable data.
Most PPC accounts have more than one ad group, so just because a marketer or business is letting data accumulate in one area of an account does not mean they are not doing anything towards the betterment of their campaigns overall. While an A/B test occurs in one area of an account, other things can usually be tweaked elsewhere that will not contribute to over-optimization.
The capability exists to make big changes fast, and sometimes those changes are obvious and need to happen immediately. However, sometimes something good might be changed before it has a chance to prove itself. What are your thoughts on how much data is enough to make a decision for PPC?