[LINK] Twitter drop-off rates (was Re: Forum on high speed bandwidth in Australia - this Thursday)

Eric Scheid eric.scheid at ironclad.net.au
Mon May 4 17:17:29 EST 2009


On 4/5/09 4:26 PM, "Stilgherrian" <stil at stilgherrian.com> wrote:

>  * People who DO start using Twitter properly soon move from
>    using the website to using client software -- which calls
>    the API directly and therefore won't be measured by
>    Neilsen's JavaScript web bugs.

What web bugs? the only ones I see on twitter.com are google analytics.

A somewhat better approach might be to look at a large sample of accounts
and note the date of their last tweet. This would of course exclude any
locked accounts, but would include all API driven posts. It wouldn't measure
how many are still simply reading (and not posting, aka "lurking").

e.



More information about the Link mailing list