Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Twitter RT Test Results (gabrielweinberg.com)
31 points by riffer on May 8, 2010 | hide | past | favorite | 13 comments


Gabriel, thanks for writing this up.

I've noticed similar results from Twitter. I've only got about 1/3rd as many followers as you on my primary account but follower count has been growing fairly steadily and organically (ie., I haven't been doing mass follows or contests or anything like that). I've had some worthwhile engagements with other users but Twitter has been terrible as a traffic generation tool.

It doesn't seem terribly hard to grow follower counts for their own sake, but I'd love to hear stories about if and how others are successful with Twitter. (Where success is measured by a metric that has value outside of the Twitter-verse.)


I RT'ed that original tweet from @duckduckgo but it doesn't show up on the twitter search page; is that because my tweets are protected? Also, I saw at least one retweet which was from someone who retweeted because of me ...

All this to say, your stats are likely a bit off ... so take it all with a pinch of salt :)


is that because my tweets are protected?

Yes. Can you imagine the furore if you could find people's protected tweets by merely doing a search? (I seem to recall, however, that they did accidentally allow this for a while when they bought the search service from Summize.)


Thx--that's weird. However, I took the internal RT counts from twitter itself (they tell you how many times it was retweeted), so if you literally hit 'retweet' I should have gotten it.

Additionally, the end metric that really matters is clicks, which I was able to count correctly, and which were of course not thrilling :(.

In any case, thanks so much for RTing!


You can see the clicks on The Denver Post's ( http://twitter.com/denverpost ) first twitter advertising push here: http://dpo.st/ipadsh+


Always use hashtags with care. While they can sometimes help a tweet make it into specialized aggregators, overuse or unnecessary usage can make a tweet look spammy or insincere.


And more importantly, the hashtag has to be recognized by your followers as something beneficial to be associated with.


I've launched a website dedicated to tracking responses on twitter. You will be surprised on the stuff that mostly gets retweeted, so I think your request was simply not interesting to the many.

My website is http://topytalk.com, it has few limitations: displaying only tweets with at least one reply, and missing about 20-40% of all tweets due to technical limits.


I said that in the post, though I'm very interested in what are the common themes in what does go viral. Can you elaborate?


Hard to sum up in a few words, but one scenario is quite noticeable: the most retweets come from a group sharing similar interests, they basically retweet just for the sake of making this interest into a trend.


Can you sum up in a lot of words :) ? Really, I think everyone would find it quite valuable.

By interested to the trend, do you mean specifically like creating a presence for a hashtag?


Maybe I'll do some sort of analysis later, also I'm working on another twitter app for the retweets specifically.

For now look at this tweet from God_Damn_Batman (7620 followers):

Invented a new drink called The Batman. Ask for vermouth, coffee liqueur & banana vodka, then crush the bartender's windpipe. So good...

It got 42 retweets in 1 hour 48 min, and I can be missing another 5-10.

It all boils down to how many of your followers think your tweet will be welcomed by their followers.


Anyone used http://140proof.com/ ? If so, results?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: