I couldn't disagree more, actually. When what you're counting is uniques, the Gawker model breaks down. You need big stories that drive out into the wider Internetsphere... Big stories require time and effort. And that's
what The Atlantic has done well. We drive engagement with regular, fast posts and we drive reach with big stories.
That said, I'm all in support of other metrics. I think time on page makes a lot of sense, but its measurement is also very inaccurate. I think comments and other types of engagement on social media are good markers of quality. But the sad thing I've encountered is that uniques get you in the door. They are the condition for being able to sell the audience's quality and engagement.