The Fallacy of Time Spent on Site

When I was working at Dow Jones, senior management put a lot of pressure on Nielsen to replace unique visitors with Time on Site as the primary metric. This was mainly because the product team at WSJ was using AJAX applications to improve the user experience instead of forcing users to click to a new page and refresh the advertising. Nielsen seemed open to the suggestion, and now I see TOS quoted all the time. But here are the major problems with it:

1. Advertisers don't care about TOS; they care about impressions, and they still calculate CPM. TOS doesn't factor into the equation.
2. TOS is skewed by users who come to your site and leave without closing their browser.
3. A high TOS metric doesn't necessarily indicate a positive user experience or high user engagement. Sometimes it means users can't find what they're looking for and won't come back.
4. TOS doesn't necessarily correlate with a conversion. This needs to be calculated for each individual site. In some cases users who spend more time on your site may be your best customers. In this case, it would be an important metric, but it only captures one aspect of a user's visit, and it should never replace the Unique Visitor count, which most accurately measures the size of your funnel.

Bit.ly offering even shorter URL's via J.mp

Silicon Alley Insider reports that Bit.ly is offering even shorter URL's for posting on twitter, reducing the number of characters by 2 and leaving more room for your tweets. The service is offered via j.mp.