One year ago, I decided to start an internal A/B experiment on the mobile-version of my website.

I've got a map, where the ad located just bellow it has the higher 'Impression RPM' of my website. I decided that:
- 50% of the ads were responsive
- 50% of the ads were 300x250 fixed

The code for each format are the following:

Responsive

<center><div style="min-height: 250px;"><ins class="adsbygoogle" style="display:inline-block;width:300px;height:250px" data-ad-client="ca-pub-12345" data-ad-slot="54321" data-ad-channel="11111" data-full-width-responsive="true"></ins></div></center>

300x250 fixed

<center><ins class="adsbygoogle" style="display:inline-block;width:300px;height:250px" data-ad-client="ca-pub-12345" data-ad-slot="54321" data-ad-channel="22222" data-full-width-responsive="false"></ins></center>

Google was constantly recommending me to implement responsive ads, and I thought the the responsive ads were to have higher 'Impression RPM' then the fixed ones. However, after 12 months, I see that the average 'Impression RPM' of the fixed ads is 6% higher than of the responsive ads. Just for the same position of the ad (bellow a map).

I'm not sure if I'm doing something wrong. Any similar experience is welcome