Expand my Community achievements bar.

Join us at Adobe Summit 2024 for the Coffee Break Q&A Live series, a unique opportunity to network with and learn from expert users, the Adobe product team, and Adobe partners in a small group, 30 minute AMA conversations.
SOLVED

Data discrepancies between Facebook Ad & Adobe Analytics reports

Avatar

Level 6

We discover huge discrepancies for our registration metric for the Facebook Ad campaigns in Adobe Analytics when comparing it to reports from Facebook Ads.

Facebook overreports by roughly 300% comapred to Adobe Analytics and compared to our database.

We tagged the links from our ads with some identifier, so when user reaches registration event, he will be attributed to this campaign tracker in Adobe.

I did some research and found that the attribution models from Facebook vary for clicks and the like. However when using custom events, shouldn't they match?

When user registers an event is triggered which is tracked in Adobe and Facebook (fbq('track', 'CompleteRegistration')

Any suggestions where this discrepancy could come from and on how to make both tools comparable?

1 Accepted Solution

Avatar

Correct answer by
Community Advisor

I have seen similar issues when looking at click throughs from Adwords banners.

What I have found is they are always often considerably disparate. They will likely never come close to matching for several reasons.

Firstly can you ask facebook do the click throughs they report account for viewable click throughs or all page click throughs? For example if an add is way below fold of a page.

Many online ads today are triggered by bots and crawlers so they are counted as a click through(by banner server) but then do not even allow your page to load completely or allow your analytics beacon to fire. Have a look as well to your page load speeds in extreme cases long page load speeds drop some of your potential conversion numbers. Also look at cookie duration of ad server, as often long cookie durations over inflate(in my opinion) traffic and click volumes of users who are actually of most relevant interest.

Another area today we see is actual browser setting of "DO NOT TRACK" being turned on. This can kill analytics cookie tracking and account for some users not being tracked as well.

In my experience your own site defined metrics should be your final KPI measure of success. So in your case ensure the KPI of a Complete Registration is setup correctly to fire, and ensure your ECID/ICIDs are well organized to allow for very granular tracking of users visits. When a real human user clicks that banner, then completes registration within a reasonable timeframe you will then have the most useful data to tell you what creatives were indeed the most successful.

good luck

View solution in original post

3 Replies

Avatar

Correct answer by
Community Advisor

I have seen similar issues when looking at click throughs from Adwords banners.

What I have found is they are always often considerably disparate. They will likely never come close to matching for several reasons.

Firstly can you ask facebook do the click throughs they report account for viewable click throughs or all page click throughs? For example if an add is way below fold of a page.

Many online ads today are triggered by bots and crawlers so they are counted as a click through(by banner server) but then do not even allow your page to load completely or allow your analytics beacon to fire. Have a look as well to your page load speeds in extreme cases long page load speeds drop some of your potential conversion numbers. Also look at cookie duration of ad server, as often long cookie durations over inflate(in my opinion) traffic and click volumes of users who are actually of most relevant interest.

Another area today we see is actual browser setting of "DO NOT TRACK" being turned on. This can kill analytics cookie tracking and account for some users not being tracked as well.

In my experience your own site defined metrics should be your final KPI measure of success. So in your case ensure the KPI of a Complete Registration is setup correctly to fire, and ensure your ECID/ICIDs are well organized to allow for very granular tracking of users visits. When a real human user clicks that banner, then completes registration within a reasonable timeframe you will then have the most useful data to tell you what creatives were indeed the most successful.

good luck

Avatar

Level 6

Very useful, thanks for sharing your experiences.

After some research I too found, that there are several difference in data collection methodology.

In the meanwhile  I found this blogpost which also shares some light (they talk about Google Analytics, but the same applies to Adobe analytics): https://supermetrics.com/blog/facebook-and-google-analytics-data-discrepancies/

Avatar

Level 1

I don't know what metrics you guys choose for analyzing facebook ads campaign but according to my knowledge if you are getting your objective on your facebook ads then here is some metrics that you should analyze that either they are running well or there is some changes required in it.

and the metrics are...
CTR
CPC

CPM
CPA

Check this to understand facebook metrics in detail: 11 Important Facebook Metrics You Should Never Ignore