Why We Decided to try Ezoic
In-house testing wasn’t getting us where we wanted to be. We had been through a period of just over 2 years where our development was focussed on A/B testing. We would meet – talk about bread crumb location, an ad position change or whatever – then decide to try a couple of variables (one against another) using a manual testing tool called Optimizely. We’d then run the tests to see if that change had an effect on the bounce rate, page views per visit or the income goal we’d set. It was a brutal process; it took a long time to get each test going and, to be honest, we got nowhere fast.
The main issue was trying to find the balance between revenue improvement and usability metrics. We’d start to make more money, but the bounce rate would go up, or we’d get page views per visit to creep up, but the revenue would tank. So, when the folks from Ezoic (a certified AdSense Partner) came knocking, we were skeptical as to whether they could achieve success where we had failed (especially since we’d done a lot of manual testing on our own) but we were also sick of trying to do this and not getting the results we wanted. I personally wanted to try it to see if we would actually gain any advantage or maybe just gain some sort of insight. We had a Google rep to talk to and he’d been very helpful, but we felt a little stuck, so we decided to dive in and give it a go. There wasn’t much downside risk if we tested just one site for a month or so.
Getting traction took about 2 weeks on the first site we tested.
Expectations were not high and after signing up. We were shown a few test layouts and went to view more of them using the Ezoic tools and, to be completely honest, we didn’t like the look of all the test layouts we saw. Some of them were okay, but a few of the desktop themes looked clunky. My team was worried that the users wouldn’t like the new layouts based on what we’d seen, and I was finding it hard to believe that this auto-testing approach would add the value Ezoic claimed. Also, it was unsettling to think that we had only seen a handful of the templates, and there were many others that were going to be shown to our visitors with our brand on them. Suffice to say, we were nervous.
After a lot of back and forth with the Ezoic team about the method of data-driven design and letting the users decide, we decided to stick with it and begin the testing process, even though it all felt a bit too much like revolution, rather than evolution.
And I’m glad we did. It took about 2 weeks (on a site with around 150k/visits/month), and I was absolutely amazed as to just how many tests the Ezoic system was getting through in such a short space of time. I could tell that the users were responding well – bounce rates improved, time on site went up, visitors were viewing more pages – all of which are positive metrics in terms of how a user interacts with your site. Also, we were beginning to see a slight lift in revenue.
Here’s the usability results from those early days from the Ezoic User Interface.
Here are the income results (numbers blurred out, but you can see the trend)
Ad Income Up Nicely.
We are now seeing a consistent 50% lift in the revenues vs previous earnings across the board for all sites. This is impressive when you consider we had a Google rep, a team of guys working on monetization, and dedicated a A/B Tester in-house. Generating this kind of extra money, while seeing more traffic and better user stats is exactly what we were hoping to do on our own, yet never could quite achieve.
Usability vs. Personal Preference
Coming to terms with the methodology of testing and letting go of our own opinions was difficult. As I mentioned, we weren’t thrilled with the original templates we previewed. However, site design isn’t all about the looks. If one thinks about Google and thinks about the simplicity of any Google site, it’s a pretty important lesson when we think about design and I think Ezoic has taken advantage of that logic. They have made the site more user-friendly and the usability of it trumps really killing ourselves over the design.
Taking that leap of faith has really paid off, and it’s something that I encourage more publishers to do if they want the best for their users.
Living with Ezoic:
One of the most interesting things we’ve witnessed while working with Ezoic is the continuous improvement. Every month or two we’ll see a new layout become the front-runner and win the majority of our traffic. Most people test until they find something better and stick with that, but that doesn’t account for changes in user behavior or how people interact with our site now compared to six months ago. Seeing the changing layouts and the continuous improvements is a clear sign that data-driven design and continuous testing is the way of the future.
So what do we do with our time now that we have effectively got a computer doing the job of an A/B tester, an ad tag management person and full time designer? Well, having been with Ezoic for well over six months and seeing what their technology is capable of, I am fully confident leaving the task of layout and ad placement optimization up to them and focusing instead on what we are best at – our content.
The testing carries on:
Brighthub.com – SemRush Keyword Ranking is up from 76k to 133k ranked keywords in one year.
Brighthub.com traffic is up by 42%, with mobile visits doubled vs last year
Keywords are up – and that of course means that traffic is up too. I compared the same period last year (February when we started with Ezoic to August using the comparison feature in Google Analytics: 2014 vs 2015)
Brighthubeducation.com – up from 62k to 95k keywords in Semrush
Brighthubpm.com – up from 20k to 34k ranked keywords
Brighthubpm – traffic up 64% (from Google Analytics)
brighthubengineering.com – up from 9k to 14k
brighthubengineering traffic is up 48%
Our EPMV (earnings per thousand visitors) growth over the first month of testing
What do you think of our results? Has anyone else had experience with Ezoic?
To learn more about the program, visit their website.