By John Cole CCO of Ezoic
Most webmasters are familiar with the concept of testing, and perhaps even testing AdSense ads in new locations, or even testing a whole new themes for their site. What exactly is a page layout testing? How do you do it? How much does it cost? What are the potential benefits?
What is page layout experimentation for content sites?
Page layout experimentation is simply the same content in new layouts. Taking the same content & reconstructing it into new (experimental) page layouts to see what works. This usually this means sending a portion of a site’s visitors to several layouts at the same time (which eliminates seasonality/time of day changes). Ecommerce sites do this all the time for a/b testing landing pages and the same methods are now being used by ad-funded publishers to boost time on site, page views, lower bounce rate and increase ad income as well. When there is enough user data to know if one layout truly beats another (the earnings per visit are higher for example), then the winning layout can then be applied to most of the traffic – boosting overall site income. So long as the traffic sources are consistent, this is the most reliable way of improving a site.
Why does this help a site improve?
There is no such thing as a perfect website. There are always better ways doing things on desktop and mobile devices. Most sites have problems that negatively affect user experience and engagement – these issues remain undetected by the site’s owner because they are blinded by familiarity. There might be poor locations for navigation element, poor color choices, wrong fonts, and bad ad positions, or a thousand combinations of things that are ‘getting in a user’s way’. Monitoring the metrics at the same time as making changes sounds obvious, but a lot of webmasters forget to test and measure. Users are silently and continually ‘voting’ (with their behavior and actions); all you have to do is put the tools in place to tap into this amazing resource and reap the benefits. User data shows the way towards better usability and improved user experience metrics have a direct effect on improved organic search rankings.
Quick example: 8,000 people are tested with a ‘left menu layout’ on desktop and 8,000 people see the same content, but with a ‘top menu layout’, with a change in banner color a different ad configuration. The content is exactly the same. The users are all arriving from a similar traffic source (organic search)… so which one do users like and use best? Do they use the navigation bar more or less? What about the ad positions on the page – does the 300×250 work better than having a 468×60 and a text ad bar? The data from the tests will soon tell you all you need to know.
How do I do this on my site?
You can do it manually by creating a/b experiments yourself using a manual multivariate testing software like Adobe test and target, Optimizely, Wingify or similar software. Or, you can use the automated layout testing platform – Ezoic.com – if you want to do whole site testing with all the experiments generated for you.
Manual testing – Pro’s & Con’s:
Pro’s: You get to keep control of all of the experiments, set them up for mobile/tablet/desktop and create the ad position and size tests yourself.
Con’s: It’s a lot of work to try and build multiple experiments to test for UX and income. Testing will get expensive if you have a lot of page views (as most manual tools charge on a cpm basis plus a monthly charge). Trying to optimize ad units and track each ad for each screensize without an API is a laborious process. As with all testing – you need to be patient to see the results.
Automated testing – Pro’s & Con’s:
Pro’s: Ezoic is a certified partner with Google AdSense and is able to automate tests for ad positions/sizes as part of overall user experience testing. All the experiments are generated automatically – for mobile/tablet & desktop with no need to change the site’s code. Integrates with CDN’s and Google’s Ad Exchange system. No monthly fees required. Claims to double AdSense income.
Con’s: Need to fully commit to the automation. As with manual testing, patience is required to see results.
Summary
Whenever you make a change to your site you should track and measure the changes in user response. Take that idea several steps forward, you’ll start to want to conduct simultaneous tests on multiple screensizes. Whether you use a manual system to try to improve your site, or use an automated system – testing scientifically will improve your site faster and will deliver permanent returns. Whether that be doubling your monthly income from AdSense or enhancing user experience for all your users, it’s a win/win for all concerned.
