1

There might be a better name than "analytics dropoff", but I'm brand new to this, and that's the best I can come up with.

There's this similar question, but it seems AdWords specific, and not nearly as big of a problem as I'm seeing.

I'm running my first PPC campaign, and it's on Twitter. Picking a random recent day, on June 7th, Twitter claims they had 1,305 clicks.

According to GA4, I had 615 Views (but apparently this counts views of multiple pages), and 459 Entrances.

What's going on with this drop-off in reported numbers? Is it normal to see ~ 65% fewer visitors in GA than in an ad platform? Is "entrances" the wrong thing to measure in GA4, are 65% of twitter clicks bots, or there is some other cause I haven't though of.

How do I diagnose this discrepancy in numbers, and what's the right way to determine traffic from a PPC campaign?

1 Answer 1

2

Not sure what LuckyOrange is, but you're making a few mistakes right away:

  1. You never make a conclusion of an ad campaign based on the relation between impressions and clicks. You always take into account your final conversion.
  2. To take the final conversion into account properly, you have to make sure you're tracking it properly, using the best tracking practices.
  3. It's best to use the same analytics solution to track users from the very beginning to the very end of their journey. Classically, the journey is from landing to the conversion.
  4. There will always be difference between the data in different analytics systems because they track differently and have different definitions of things.
  5. The difference between clicks and pageviews will always be different, but it's best to keep the difference as low as possible. A lot of things contribute to the difference: starting from donor sites putting ads in places where users are likely to misclick and finishing with the recipient site having rendering/tracking/timing problems.
  6. In GA, you want to look at pageviews where the page is your ads landing page and users source and medium indicate your campaign. You compare that to your ad clicks.
3
  • Thanks for this. Would love to have a discussion on all of your points, but this doesn't seem the place. I've updated the question based on #6. Do you have a link for #2? Jun 9 at 17:12
  • 1
    2. without knowing what your conversion is, it's quite hard for me to say in particular, but what I meant when I wrote that was EEC. But hey, if EEC is not applicable, so, say, your conversion is email subscription, then still, you have to track the subscriptions properly (success, error, attempts, etc) and make sure you can build a funnel from the landing to the conversion to make analysis and reporting much easier. You can always join Measure slack and ask there. Or Libera IRC, and have a discussion there. linkedin mb? I would say measure is the strongest, bit it will take time to join.
    – BNazaruk
    Jun 9 at 20:36
  • about the data to twitter clicks discrepancy, I suggest you test your tracking. Make sure it fires properly every time. Don't look at the entrances, they're not super useful. Take a look at the number of pageviews generated on your landing page with twitter as a source/medium.
    – BNazaruk
    Jun 9 at 20:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.