It's true that for NPR the effects have been negligible (overall traffic dropped by 1%). However, this is mostly a function of the fact that less than 2% of NP

It's true that for NPR the effects have been negligible (overall traffic dropped by 1%). However, this is mostly a function of the fact that less than 2% of NPR's traffic came from Twitter in the first place. Even if it lost all of its traffic from Twitter, the effect could still have been described as "negligible". But when you look at this in terms of percentages, NPR's traffic from Twitter dropped by over 50% after it deleted its account. So another publisher that's considering making this move should expect to see its Twitter-sourced traffic to drop by a similar percentage.