optimize google adsense experiments

Google AdSense Optimizations

Once you have started earning some money with Google AdSense, you may get some notifications about various optimizations that you can do to “improve your earnings.” Do the Google AdSense optimizations really work? It depends upon the optimization and how it applies to your website.

Google Optimize Suggestions

Remember, that Google works best for Google. Those search results that are scraped information from other websites that appear on the search results page? That helps Google. The ads above the fold, and links to Google properties. That helps Google.

That is true for optimizing AdSense as well.

optimize google adsense experiments

The good news is that as an AdSense publisher, you get a cut of what Google earns putting ads on your website, so that puts you on the same team. However, that doesn’t necessarily make what Google suggests the best for you. It doesn’t mean that they are trying to reduce your earnings either. Remember that Google has an enormous number of websites to fill with ads. They think in the aggregate. You, however, think about what is best for your specific website.

For example, I have a freelance writing blog that I publish with tips and information for both freelance writers, and clients of freelance writers. The people this website are directed at are people doing freelance writing — or wanting to do freelance writing — now, which means that ads for colleges or other writing programs aren’t very effective on this website. Unfortunately, ads for writing courses were essentially filling up every ad slot on that blog.

This isn’t Google’s fault. The website is an educational website for writers, after all. However, the kinds of people reading it, generally won’t be the people looking to go to school to be writers, especially at the sketchy kinds of institutions that often show up in those particular AdSense ads.

I turned those ads off. Google allows you to restrict certain ads, or even ads from certain domains. Ironically, they don’t really provide this opportunity in order to let you maximize profits, so much as to prevent competitors from advertising on your own websites. For example, if I ran writing school, I wouldn’t want other writing schools publishing ads on my website. However, you can use the blocked ad categories as you see fit, and I block the Education & Training category. I blocked Ringtones and Downloads on my personal finance blog because those took a lot of impressions for little returns. (It makes sense that people looking to improve their finances aren’t dropping money on costly ringtones, right?)

This bugs Google AdSense because it knows that overall, on its millions of sites, that these ads make money. But, you know your website better than Google, so should you trust your gut, or go with Google’s non-specific data?

Google AdSense Experiments

Fortunately, you don’t have to choose blindly. AdSense allows you to run optimization experiments. What it does is run half your impressions with those ads blocked, and the other half with them unblocked, and then it reports back how it affects your earnings, including things like number of clicks and cost per click. It does take a few days to fully run an experiment, but you can usually tell which way the data is going pretty quickly.

In my case, blocking the Education and Training category performed much better (+48%) than not blocking it. So, I kept it blocked.

However, it turns out unblocking the other subcategory, Job Listings, actually improved my earnings. I’m not in this for purity, so the block for those ads came off.

Click the Optimization tab from time to time to see what, if anything Google suggests. When they do make a suggestion, click that Experiments link and check it out. You may even want to repeat the experiment later on, especially if you writing focus changes.

Share and Enjoy

  • Facebook
  • Twitter
  • Delicious
  • LinkedIn
  • StumbleUpon
  • Add to favorites
  • Email
  • RSS

Leave a Reply

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