I know the role of featured images is to grab attention and draw readers in, but I don’t like them. I particularly hate humongous ones that occupy a large chunk of the page and require several hours of scrolling (maybe) before you actually reach the content.
From a publisher’s point of view, I don’t like them because it’s too time consuming and difficult sourcing suitable images to plonk at the top of a post. Yeah, it’s fine if you have access to a photo agency, but for most people it’s either a case of trawling through Flickr for pics, or, as most small-time bloggers tend to do, nab whatever you can find via a Google Image search and hope the creator never finds out.
Unfortunately, I think I am in the minority when it comes to my dislike of featured images – please, someone tell me they share my view – so I have been reluctant to remove them from this, my little corner of the internet.
What’s that? You can’t see any featured images on this or any other post on here? That’s because I am currently conducting an AB test to analyse user engagement when they are removed, so you’re in the group that won’t see them for the next month or so.
It will be interesting to see what impact – if any – their removal has. The three things I will be comparing are page views, session duration and Adsense revenue, because I need extra pocket money to buy more sweets and fizzy pop.
So expect to see a follow-up post at the end of April with my findings. I may even sprinkle a few graphs in it too, but in the meantime, if you are wondering how to conduct an AB test on your site, then continue reading. If you’re not, then I’ll hopefully see you again in four weeks’ time.
AB testing using Google Optimize
There are several website testing tools that allow you to conduct an AB test, but the easiest method I found was to use Google Optimize. The only problem, however, is that it is relatively new and may not be available everywhere at the moment, but you can always use Google Tag Manager instead.
Update Google Optimize is now available in over 180 countries. For more details, check out this post on the Google Analytics blog.
Optimize is easier to implement and a doddle to use. It also does not exhibit what is known as flash of original content (FOOC), which is a noticeable flicker of the original version of a site before test changes are seen.
This tends to happen with Tag Manager due to the script being loaded asynchronously and not executing soon enough. I guess there are workarounds, but I gave up after 30 minutes of trying to find a solution.
If you want to know how to use Optimize for AB testing, then I suggest you watch the following video by Julian Juenemann over at Measureschool. Like all his tutorials, it is excellent and I would encourage you to subscribe to his YouTube channel immediately.
As I previously mentioned, if you don’t have access to Optimize at the moment, then you can do a split-test using Tag Manager, although it is a little tricky getting it to work, but once again Julian has produced a fantastic guide for you to follow.
Admittedly, I had difficulty receiving and filtering the data in Google Analytics using the Tag Manager method. This coupled with the FOOC issue – and my lack of patience – is what lead me to use Optimize instead and I’m glad I did, because it is a great site tweaking tool, even at this early stage.