Part 3: Facebook vs. LinkedIn vs. Adello Direct mid-term performance analysis

We are almost half way through our live competition between Adello Direct and the other big two. As I promised when starting the competition we’ll be transparent and show our targeting funnel as well as the performance outcome in a continious way without polishing up behind the fences.

Unfair advantage or unequal competition?

So for me as the Product Marketing Manager behind Adello Direct Mobile Marketing Cloud this approach comes with the huge risk of failure. First, the risk of failure by choosing the wrong targeting options on the three different platforms. And second, because the competition could potentially result in very poor own performance outcome of Adello Direct. And third: if our performance results will be better, I still run into the risk you’ll not believe me.

As for the first risk: yes, I can clearly see that argument and I already partly discussed that openely in my previous blog post about the targeting parameters for a B2B advertising campaign. Finding the right targeting options that create a more or less equal campaign on three platforms cannot be achieved to a 100% satisfaction. At least not to my very own standards and definitely not on a scientific level either. Some targeting options on Facebook are more precise in its targeting funnel than on LinkedIn whilst other campaign features on LinkedIn are better than on Adello Direct and vice versa. So, whoever wants to challenge me here: swing the bat – you are right. Difficult to compare, you have to find your very own platform what will work best for your B2B ad campaigns.

But what about the performance outcome? Remember: we have a B2B branding campaign, thus are aiming for high results when it comes to CPM, so we’ll judge by the overall impressions as well as by the click rates (if possible).
So, here are the numbers after 24 hours and after seven days in the competition:

Total number of Impressions and Number of Clicks

After 24 hours:
Facebook: 14.708
LinkedIn: 1.713 number of clicks 3 -> CTR n.a.
Adello Direct: 61.034 number of clicks 88 -> CTR 0,13-0,18% (the overall summary of the click rate shows a bug, but the individual flight CTR rates are correct)

Facebook Branding campaign results after 24 hour
LinkedIn Branding Campaign results after 24 hours
Adello Direct Branding Campaign results after 24 hours

After seven days*:
Facebook: 56.020 number of clicks:
LinkedIn: 13.053 number of clicks: 28 -> CTR of 0,21%
Adello Direct: 170.210 number of clicks: 297 -> CTR of 0,15-0,19% (the overall summary of the click rate shows a bug, but the individual flight CTR rates are correct)

*we had to pause all three campaigns on August 19th due to accounting issues. The campaigns were all reactivated after the issue was solved. That explains the gap on the Facebook graph below

Facebook Branding campaign results after seven days
LinkedIn Branding Campaign results after seven days
Adello Direct Branding Campaign results after seven days

Whilst the performance output fairly differs, the performance outcome seems to speak for itself. However, it gets even more interesting when looking more closely into the performance results data we are provided with by Facebook. Facebook provides us with a reporting file in which they list all the inventory pages that the ad was placed on.

Thumbs up for reporting transparency on Facebook, but…

One has to acknowledge that the type of transparency gets my two thumbs up. But checking back with my targeting settings for the campaign which decided to go for German speaking audience only, I somehow wonder why I see quite some inventory in foreign language inventory.

Language targeting for German speaking audience on Facebook

Above for reference the targeting of the campaign. And below an abstract of the inventory used by Facebook.

Some examples of the inventory Facebook used to place our ad

Yes, maybe these pages do get a lot of german speaking audience, but the amount of inventory clearly being non-german seems to be overwhelming.

Yes, programmatic advertising means to have a mix of different inventory, but really, that much??

So for cross-reference, I check back with Aaron Salzgeber who is the Head of our Ad Operations team at Adello and tell him about my findings and his opinion. This is what he shared with me:

There is a lot of traffic from foreign sites within Germany and Switzerland available, also on our exchanges. We try to keep them out by not whitelisting them (only english sites and sites in the local language, however since the sites are worldwide whitelisted sometimes we also have inventory from APAC where we serve). However, we saw that if something foreign is whitelisted the same happens in our network as well…(also with language targeting)

Aaron Salzgeber, Head of Operations at Adello AG

My mid-term findings are the following:

A high performance number in impressions does not guarantee any performance results in your targeted audience group

A closer look behind the scenes does make sense, but is not being provided by all platforms

We have to take into account that the campaign itself is not comparable from a scientific point of view.

To all the Marketing Performance Managers out there, I raise the question: do you take a look behind the performance numbers and yes: what are your findings? Happy to discuss your point of view!



Leave a Reply

Your email address will not be published. Required fields are marked *