It's easy to understand both the reason this could be true, and the reason not to care. When someone visits your website from SU, they just clicked the stumble

It's easy to understand both the reason this could be true, and the reason not to care. When someone visits your website from SU, they just clicked the stumble button. How many pages have they viewed in that session? How many of them bounce immediately? How many of them read a page and - while in their stumble mode - stumble along to the next page? And when someone visits your website from FB, they probably just clicked on a link a friend shared with them (implicitly endorsing you), or came from your FB page, where they have been pre-qualified as interested. As other commenters have pointed out, pageviews aren't the best way to measure the efficacy of your digital promotions. Try your best to find metrics that align with your actual business goals. Conversions to sales would be a good one, but something like time spent on site or pages viewed per session are fair indicators that you're doing something right. I'm not trying to decry SU as worthless or this article as baseless; not by any means. I just want to encourage a healthy skepticism of metrics that don't necessarily mean anything. If "that which gets measured, gets managed", then optimizing for page views could mean you find yourself in the deadpool soon.