Google Analytics "Ghost" Transactions

Options
Hi,


We are tracking transactions through Google Analytics eCommerce and see every now and then a "ghost" huge transaction of $50K and over, which is not a real transaction as we don't see it in our LO database, there's no record of these transactions. 95-96% of the other transactions do correspond to the transactions in LO, except for those huge ones. In Google Analyitcs the transaction shows as amount only, but no Quantity, it lists “0” under Quantity, all the rest legit transactions, have “1” under quantity.


Has anyone experienced anything similar?
Tagged:

Comments

  • Hi Anita -


    For the most part, that sounds like someone is testing scripts or posting to your UA instead of their own.

    I would look at the traffic source information and transaction id - if they're consistent then one option would be to filter that traffic out. 

    If you have Donations set up as a Goal in addition to just eCommerce, you can take a look at Goal URL and see if it's consistent there. 

    Additionally, you could create an advanced segment for anything uniquely consistent and poke into specifics to either further identify or block this occurance.


    Another fun check is in the Donation Campaign settings for maximum gift amount as well as in the Site Data Parameters (SDPs) for the same. If those are set below what's being reported and collected by GA, then you definitely have somebody using your code.


    And as always, a friendly reminder that Luminate Online, your database of record or other transactional systems will never match a web tracking tool 100%, ever.


    Happy coding,

    Sara



    Sara Hoffman

    Digital Analyst and Optimization Lead

    Beaconfire RED
  • @Anita Yankova Your Google Analytics might be tracking the “Test Drive” transactions as well. Generally, when you test drive a donation form, the transaction ID is always zero which you can filter out. You can configure your Google Analytics tracking script to not track any of the transactions done with “Preview” parameter in the URL for cleaner data.

Categories