After just 0.1 second users start to perceive a delay if your app isn’t doing The Thing™ that they just told it to do. We needed to buy 8.5 seconds. We had to manipulate (percieved) time.
Are ghost buttons really as bad as they've been made out to be? Ghost buttons are often touted as low-affordance, and it’s true of many of the examples we see — buttons with poor contrast placed over images making them difficult to use and confounds A/B tests. We say we are comparing ghost buttons and ‘normal’ buttons but we are really testing accessible vs inaccessible buttons, high vs low contrast designs, high affordance vs low affordance designs (and it’s not a surprise that ghost buttons lose). What happens when we test an accessible ghost button against an accessible solid button?