Register & Login

SociafyQ Forums » SociafyQ - Help: Questions and Answers

Twitter Retweet Rank

(8 posts)
  • Started 1 year ago by rjleaman
  • Latest reply from therendStudio

Tags:

  1. rjleaman
    Member

    Can you explain what goes into the Twitter Retweet Rank?
    I get it that "lower is better" but can't seem to figure out how it's calculated just by looking at the stats - which would be helpful in deciding how to interpret changes in that measurement.

    Posted 1 year ago #
  2. therendStudio
    Moderator

    Hi ;)
    Sorry for the response delay but the forum is not yet configured to send alerts as necessary.

    About retweet:
    Your retweet rank for example is

    http://www.retweetrank.com/rjleaman

    (I'm not sure what it means though :)

    In future there will be other and more useful twitter statistics fetched through its api.

    Posted 1 year ago #
  3. Hey there,

    Retweet ranking is, in the end, not as useful as we thought it might be. The basic idea (as far as I can glean) was to rank the most popular retweet keywords/pages, so one could get a sense for how much a brand was (or wasn't) retweeted in comparison to others.

    The problem is, the raw number of tweeters is constantly changing - so the rank is constantly in flux, making the ranking far less important.

    The more relevant number is how MANY retweets an individual person has; an increase in that number would be a great thing; and better, the number of individuals who are retweeting one's tweets -- yes?

    --Dave

    Posted 1 year ago #
  4. therendStudio
    Moderator

    1. The retweet ranking on that site seems to be targeting (as is the definition of retweet the way I understand it) the particular user retweets rather than keywords so that one estimates his own success.

    2. A problem with www.retweetrank.com itself I just spotted is it doesn't even accurately count actual retweets.
    For eg.
    "nsmcnahb: RT @builderonline @DeirdreReid: Great advice - 5 Tips to Revive a Fading Facebook Group (by @rjleaman) - http://bit.ly/PuKGI"

    is not actually a retweet to @rjleaman but rather to @builderonline & @DeirdreReid

    3. And yes as Dave noted while the retweet rank can show you your success compared to others it doesn't provide a clue as to what your own effectivity (and progress) is.

    So I suggest:
    The retweet rank on SQ to be calculates as

    100 * tweeter updates / retweet rank

    Thus value of 100 on the chars meaning 100 % of your updates have been retweeted once
    A value of 1000 meaning averagely 10 followers have retweeted every your update.

    Not sure if the actual follower count shouldn't be involved in the calculation though...
    Perhaps it should be left for manual calculation through the excel feeds ?

    Daniel

    Posted 1 year ago #
  5. rjleaman
    Member

    For what it's worth, I think the most useful stat regarding RTs is how many times an update that orginated with me has been repeated (retweeted) by others. It's a rough indicator of what followers are finding most interesting - but very rough indeed, as you've noticed.

    The example "nsmcnahb: RT @builderonline @DeirdreReid: Great advice - 5 Tips to Revive a Fading Facebook Group (by @rjleaman) - http://bit.ly/PuKGI" is a good one to show the inherent difficulty with this metric. (Did I originally tweet that link to my post, and then it was picked up and paraphrased by others? Or did the tweet originate with @DeirdreReid? I can't even remember, but it could have been either way.)

    I'm inclined to agree with Dave that "The more relevant number is how MANY retweets an individual person has; an increase in that number would be a great thing; and better, the number of individuals who are retweeting one's tweets -- yes?" It would still be only partly accurate, because of the nature of language and RTs, but volume of RTs would be a useful tracker and it would also be useful to see if all the RTs are coming from the same few individuals or whether the "reach" is increasing over time.

    Hope that makes sense!

    Posted 1 year ago #
  6. therendStudio
    Moderator

    and it would also be useful to see if all the RTs are coming from the same few individuals or whether the "reach" is increasing over time

    Great tip! *

    I suppose eventually (when the scheme be somewhat clarified) the retweets will be a separate tracker type showing something like:

    1. Absolute count of own retweets
    2. Percentage of retweets to one's updates
    3. Percentage of retweeters to one's followers *
    4. A mysterious combination formula between own retweets, the unique retweeters and the followers count to give some more organic clue of the outreach.
    5. ...

    Will need a comprehensive discussion on this one though...

    Posted 1 year ago #
  7. Those sound tremendous: #'s 1-3 are key; it's the combinations that I shy from. There are quite a few sites that develop some kind of hybrid number that ends up being somewhat meaningless; the clean distillations (#1-3) are far more powerful.

    Posted 1 year ago #
  8. therendStudio
    Moderator

    Since even 1-3 are anyway combinations of retweets and stats already collected from another tracker I was thinking of adding formula (or even custom) stat entries.
    Those will show in the stats dropdown but will not need a separate entry in the db, instead will be calculated from other stats in some way. Thus any number of additional stats can be added to a tracker showing various tendencies.
    But perhaps all those should better be left for the still to be developed chart sheet pages...

    Posted 1 year ago #

RSS feed for this topic

Reply

You must log in to post.