How to Test Your Google AdSense Ads to Get the Most Revenue
Google AdSense is the biggest display advertising network in the world and is used on millions of websites. It's a great way for writers, bloggers, content creators and publishers to monetize their content; when people click on an ad that you're displaying, Google pay you close to 70% of what the advertiser paid them for that click.
Maximizing website or blog revenue through Google AdSense needn't be complicated. This simple, step-by-step guide lets you know exactly what you need to do to test AdSense ads on their website. This will help them earn the most money in an effective way, using data and repeatable techniques and decisions that will create the best click-through rates and create the most income for you.
The other thing Google AdSense has is a lot of options for how and where to display ads, what media to use, sizes, colors etc. How do you go about testing all of these options so that you can get the most revenue from having the ads on your site? This simple, step-by-step guide will show you:
- Establish what and why you are going to test ads
- Realize that testing AdSense will take some time
- Establish how you are going to test your ads
- Set up all of the different ad units that you want to test
- Setup URL channels
- Place ad units on pages
- Let your test run
- Continue testing, tweak and repeat
Benefits of testing your ads in AdSense
- Allows you to analyze what ads are performing well
- Lets you tweak your ads to make them relevant and appealing to your visitors
- Increases the possible revenue you can make from your website through AdSense
Step 1 - Establish what and why you are going to test your ads
This process is designed for publishers that want to learn what size, positioning, media and types of AdSense ads deliver the best click-through rate (CTR) and revenue (revenue per thousand impressions or RPM). We'll achieve this by creating a number of different variations on ads and placing them on your web pages to see which ones achieve the best results with your visitors.
So, for example, you could try two banner ads and one square ad on one page and test that against three square ads on another page; or you could test all images against all text or any variation that you can think of. You will, of course, need an AdSense account and you can find out how to do that by searching Google for 'create an AdSense account'.
Step 2 - Realize that getting results you can use when you test AdSense will take some time and a number of impressions
The first thing to know is that testing what works in AdSense is not a quick process. Because you need a decent number of visitors, impressions and clicks, depending on the traffic to your website, it could take a few weeks to get the results that you need.
Despite this, it will be time well spent. Having the right quality and quantity of information on how your ads are performing will definitely benefit you in the long run. Ideally, you're looking for a thousand or so page impressions (at least) on one variation of ads (and the more the better).
Please also note that the steps listed here are different to the built in 'Experiment' feature in AdSense, which allows you to test variations within the ads themselves.
Step 3 - Establish how you are going to test your ads
There are two main ways you can test variations of your ads and positioning, depending on what you prefer (and what the technical limitations of your website are):
- Option A - You can test different variations of ads on different pages at the same time or
- Option B - You can test the same variation across your whole site and when that is completed, test the next variation
You might use option A if you already get a lot of visitors to a number of different pages and want to speed up your testing so that you can identify well-performing ads quickly.
You might use option B if you get a number of visitors to your website as a whole but relatively few to individual pages; you might also be limited to option B if you are using a website that only enables you to use ads in the same positions across all of your pages (e.g. via a template).
Regardless of which you choose, the following steps will work for you.
Step 4 - Set up all of the different ad units that you want to test
- Go to Google AdSense > My Ads > Create New Ad Unit
- Create all of the different ad units that you want to test, there are three main variations you can change in the Ad Units themselves, they are:
- Size of the ad
- Media used (Text, image, combination)
- Ad style (Colors, font etc)
- There is one further variation that you can influence when you add the ad to your web pages, which is the placement on that page
Name the ad unit so that it contains all of this information, an example might be: '336x280 Text Bl/Gr Top Left' - This would mean: Size of 336x280, Text only, Blue / Green Color Scheme, placement on Top Left of site. You should continue and create variations of every ad that you want to test; this may result in a lot of ads, but again, this time will be well spent.
Step 5 - Setup URL channels
You will now want to setup URL Channels - These are channels for each of your pages in AdSense that will allow you to track how each variation on your pages is performing:
- Go to AdSense > My Ads > URL Channels > + New URL Channel
- Create a URL channel for your site as a whole (the top level domain)
- Create a separate URL channel for every web page that you want to track ad performance on
Step 6 - Place ad units on pages
The next step is to place your ads on your individual web pages. You are allowed up to three separate 'Ad Blocks' (i.e. the actual text / image ads themselves) and three 'Link Units' (auto-populated by Google that provide links to other pages outside your site where ads are shown) on a page:
- Decide on the page where you want to place the ads and link units
- Add the ads to the page, paying attention to how you named the ads and placement
- You might want to note down (e.g. in a spreadsheet or on a document) the pages where you placed ads, the type of ad that you placed etc. This will be very helpful later when comparing performance
- Repeat this for all of the pages that you want to check, either using a different variation on each page (option A) or the same variation across the site (option B)
- If you use option A, I would suggest not trying out more than two or three variations across your site at a time as any more than this will make analysis more difficult
You can create variations however you want - You could have a variation on positions within the page (banners at top or bottom), ad sizes (square vs. rectangular vs. skyscraper for example), colors of ads, media (text vs image) or anything else you can think of.
Step 7 - Let your test run
Now's the time when you need to sit back and let your testing run. You'll need each page that you are testing to accrue a decent number of impressions so that you know that you aren't basing decisions on a small number of impressions or clicks. It is recommended that you have at least 1,000 impressions for each variation that you are testing, and more if possible (the more the better).
It's also important that each variation you are testing should have roughly the same number of impressions, so that you know you are comparing the behaviour of a similar number of visitors.
You can check how many impressions your ads are receiving on the 'Performance Reports' tab of your AdSense account. In other words, you don't want your data and analysis getting skewed by a few errant clicks.
Step 8 - Review performance
Once you have enough impressions, it's time to analyze your results. The easiest way to do this is in a spreadsheet, but you can also use a document, pen and paper or whatever works for you. The most important thing is that you are able to compare and have a record of all the tests that you have done.
You can get all of the information you need from the 'Performance Report' section of AdSense:
- Go to AdSense > Performance Reports
- For each of the reports, you want to look at the specific time that the ads have been running
- The main two measures you are looking for are:
- CTR - How often your ad is getting clicked on
- RPM - The revenue you are getting for a thousand impressions of the ad (RPM is actually a better measure than most of the others because it takes into account many aspects of your ad's performance to tell you how much it has earned and likely will earn in the future)
- The performance reports can be used as follows:
- URL Channels (across your site) - The site as a whole will tell you how all of the ads across your site have been performing
- URL Channels (individual pages) - This will tell you how the ad variations on each page performed
- Platforms - This will show you how well your ads performed on desktop, tablets and mobile devices (some ads perform better on some platforms)
- Ad Units - This will tell you how each of the ad units you created and named performed across all of the pages they were placed on
- Ad Sizes - Tells you how well each particular size of ad performed, regardless of media or placement
- Ad Types - Lets you know how text performed vs. media, images etc.
You can also report against multiple dimensions in one report, so you could have the 'Ad Size' as the main measure and the 'Ad Type' as the next.
What you want to do is to note down all of the best performing ads and placements on the page, i.e.
- Which size ad performed the best?
- What was the top placement for that ad on the web page?
- What type of ad performed best?
- What web pages performed best? (This will help indicate what collection of ads work well together)
Note down all of this information, so that you can compare it, both to itself and to the next test that you run. If you're following option A, you can compare multiple pages right now and see what does well. If you are following option B, note down all of the information you need before running your next test.
Step 9 - Continue testing, tweak and repeat
Now that you've completed one round of testing and identified your top performers, you'll want to repeat the testing with different variation(s). To do this, go back to step 6 and change the ads, placements, types etc. that you need to. Then, let that test run for a while, review your performance reports and get more data.
Eventually, after running several tests, you will get a clear idea of the sizes, types and placement of ads that work best for your website and audience. Once you have that information, you can apply that variation of ads across your whole site and increase your revenue.
A word of warning
To keep your AdSense account in good standing and to prevent it being banned, you must read and abide by AdSense's terms of service. In particular:
- Never click on your own ads
- Never encourage people to click on your ads
- Never use dishonest practices to fool people into clicking on your ads
- Never put ads on copyrighted material that you don't own the rights to
- And lots more - Please do read the terms of service as breaching it will get your account banned (and there's no easy way to get another one)
Setting up Google AdSense
How do you feel Google AdSense performs for you?
Putting these steps into practice can significantly increase the overall money that you earn from AdSense, after a couple of months of testing.
A methodical approach, combined with good data and well thought-out decisions will create better revenue and success for your website.