Traffic Sources: Who is wrong? Facebook or Discourse?

During the last two weeks or so, I tried a little Facebook promotion aimed at getting people to follow the link to my discourse forum. According to facebook, 48 people clicked the link. However, discourse tells me on the admin dashboard under “Top Traffic Sources (Last 30 Days)” that I have received a total of 7 clicks from facebook (3 from, and 2 each from and Who is right?

I also checked google analytics which saw a total of 19 referrals from facebook during the past 30 days and 8 since the camaign started.

I read this somewhere a while ago but Facebook increments a counter for outbound traffic metrics before the browser loads the referred site, so if the user is quick to hit back/stop in their browser, the counts between Facebook’s outbound and your site’s inbound will diverge.

The conclusion in that particular article was that users were misclicking, and stopping before the traffic was actually received.

I cannot for the life of me remember which article it was so take it with a grain of salt, but I would personally put more weight in metrics that are counting inbound traffic.


What about mobile apps?

That makes sense though it seems unlikely that 75% of all clicks that FB registered were misclicks in that sense, i.e. that they were not only misclicks but people reacted extremely fast to “undo” them. And even that were the case, I’d expect that FB doesn’t count these, since what they’re selling to me are “Website clicks”.

For me, the conclusion is: don’t trust Facebook Website promotions and multiply whatever price estimate they give you by four in order to get a realistic estimate.