Start a new topic
Not Taken

split test social appearance


this is a big one for me.

Currently, we only have one social appearance option per campaign and yet if this is to truly be a traffic getting service, shouldn't we split test which post setup has the best conversions ?

This is a big deal I think.

This is something I've been thinking about a LOT, actually. Truth is, I'd LOVE to have this feature - but I don't see any option to split-test this in a way it makes sense.


Let's say you're split-testing 2 social appearances. 50% of the users will post version A (when they share), 50% of the users will post version B. 

Even if version A would get much more traffic or a better conversion rate, it doesn't tell us anything (we can't draw any conclusions on that). Why not? 

  • We don't know how many "friends" each person has, e.g. his audience size. If someone with a huge network shares A, than A might come out as a winner (but that's based on nothing).
  • We don't know the relationship strength of each person. If someone with a HUGE credibility shares version A, the sign-up rate from A would be better - although that doesn't mean version A is actually better.
We simply don't know the "views" of each message, we can only count how many people clicked on it... and that on itself isn't enough to draw any conclusions on.

However... I'm very open to suggestions / ideas. Please share below :) 

1 person likes this

hmmm... I see your point.

Login or Signup to post a comment