Major Discrepancy - Measured GA Sessions vs Twitter Clicks - How to Fix??

Discussion in 'Social Networking Sites' started by RiskSlayer, Sep 7, 2015.

  1. RiskSlayer

    RiskSlayer Junior Member

    Jun 17, 2015
    Likes Received:
    Social Media Agency
    Washington DC
    Hey Guys,

    Recently, I got 2,000 clicks in a day on my Twitter acct posted links (according to Google URL shortener & HasOffers click attribution)

    The links go to a 3rd party website. I cannot edit this site, but I do have access to GA on this site. GA only saw about 600 new sessions from the utm_source used with these links. That is a 70% discrepancy.

    I spent many hours researching how TW referrals are sometimes removed depending on end user browser settings and/or the particular end user twitter app. To address this, I manually create tracking links with appended parameters via Google URL builder. As suggested by my Internet research and my experiments this approach appears to effectively attribute in GA. So, this does not appear to be the reason why there is a discrepancy.

    From my experience with MyLikes and ChaCha (TW Ad networks) - I know they are seeing much closer measurements of clicks to GA sessions - while I am seeing like a 70% discrepancy. I control the TW accts, and I am not doing any blackhat stuff with those TW links (and don't believe anyone else would be faking clicks on those links either)

    Twitter is mostly mobile traffic, so I think mobile users are just exiting maybe before the page fully renders (the 3rd party website is mobile-optimized, but is still taking 10-30 sec to fully render depending on the network connection). Friends have suggested that MyLikes is persisting the click/session in some manner to score a GA session??

    I believe that MyLikes and ChaCha and others are doing something to make sure that almost all clicks they acquire are converted into a GA measured session?

    You guys have any ideas on what could be done or what Ad networks are doing on mobile to address this?