You can get all my newest conversion test results by subscribing to my RSS feed. Speak to you soon!
Ok… here we go… this is the first in a series of A/B split tests for “The Big Yellow Book of Turbo Marketing Secrets” squeeze page.
As the results come in I’ll keep you posted (here’s the first screen shot)…

(Screen shot from Google Website Optimizer)
Just in case you’re wondering what I’m testing…
I’m testing if having “as seen on” logos (A.K.A. “flares”) effects conversions in any way. I’ve heard that including flares in your squeeze page can increase conversions by 10-40% (if not more).
Here are the test pages…
Original (control):
BigMarketingSecretsBook.com
Test page with the logo “flare”:
BigMarketingSecretsBook.com/version3/index.html
Let’s see how it goes! It’s what testing is all about.
The traffic source for this page will be exit traffic from another site I run. These visitors are newbie business opportunity seekers who are looking for additional online income so it’s fairly targeted.
I’ve also got Google Analytics wired up to the test pages so I’ll be keeping a close eye on those results too.
Before I forget… make sure you subscribe to the RSS feed of this blog either through your feed reader or even by email. This way you’ll get to see my results as soon as they’re ready.
I’ll be back in a couple of days time with an update on my testing.
Remember...-You can't improve what you don't measure!
Last 5 posts...
- Big Marketing Secrets Book: Update #4
- How To Use Google Website Optimizer...
- AWeber "live" conversion stats video update...
- About to run a promo to a 7K list!
- Big Marketing Secrets Book: Update #3

{ 1 trackback }
{ 5 comments… read them below or add one }
Thanx for the good articles you write. I find this one really useful because I have heard of testing but never knew how to do it. Can it be don in the autoresponder as well?
Cheers,
Hello Gunvi
Thanks for your message. Yes it can be done with autoresponders. The only one I know that can effectively split-test email campaigns for you is Aweber, however you do need to have a list of at least 1,000 people before doing this to get any kind of meaningful results.
The other way of split testing with autoresponders is to set up rotating links using a link tracking online application like AdTrackz. This way you could split test different landing pages from the same email message. It’s more connected with general website split testing but uses the traffic generated from email autoresonders to generate the traffic.
I hope this helps.
This is a good article for advanced marketers. Btw there has already been carried out extensive split testing on landing pages and stuff. Ryan Diess has done a split testing experiment where he testing 43 different details in a landing page. Read that and tweak your landing page. For any Newbie internet marketers out there, read through the following article to get an idea of what split testing is.
Split Testing Article
Hello Hari, I know Ryan’s 43 split tests product very well and if you can afford the $97 a month price tag then it’s a great investment to make. However despite the fact that he has done this already there’s no reason why others can’t publish their results too.
There’s always more to learn and even more to test to always be improving your results.
You said I should read 43 Split Tests and tweak my landing page, that’s exactly what I’m doing. How about you? What are you split testing right now?
The problem with many marketers is they don’t do any split testing. They tell others they should be testing but amazingly just don’t do it themselves never mind sharing their results.
The fact is testing makes it possible for you to massively increase your profits without generating any additional traffic (this includes spending money on advertising).
Just in case you are a newbie here’s what split testing is all about…
Let’s suppose you have a website with a headline at the top of the page.
Your website is getting 100 visits a day and only 1 person orders from your website.
This gives you a 1% conversion rate.
Now if you change your headline it might be possible to increase your conversion to 2% (double your sales).
The problem is how are you going to measure this?
Are you just going to count up how many visitors you get then divide this by the number of sales to get your new conversion % for your new headline? Then repeat this process over and over again for each new headline you test?
Or are you going to use a service like Google Website Optimizer to manage the testing and the tracking of your different headlines?
You see what you also need to understand is that not all traffic is created equal. This will have a huge impact on your conversion % especially if you’re using different sources of traffic from different places. Each week the quality of your traffic will be different.
When you use an online system like Google Website Optimizer you can actually test each headline at the same time instead of waiting for a week before you manually change the headline again.
Google Website Optimizer ensures you see like for like results from the same traffic sources each and every day.
The best way of testing a headline is to run what’s called an A/B split test. Basically this is headline A testing against headline B.
Google Website Optimizer will tell you which of these headlines converts the most number of visitors into customers. When you have a winning headline this then becomes your “control”. You then write another headline and test this against your control in an A/B split test.
You just keep on running A/B split tests until you can’t increase your conversions any more. Then you run A/B split testing on other parts of your website to see what effect they have on your conversions.
Testing your websites headline is always a great place to start.
I highly recommend watching this video then creating a free Google Website Optimizer account…
How to use Google Website Optimizer video
Like I have mentioned in my post, keeping the context same for both the headlines is very important. Unless you do that, there is no common basepoint for you to extrapolate your results from. I prefer to use rotate atleast 3 variables at anytime for a period of 15 days. I then try to deduce which of them is the best. Any keeping the source of traffic even for the whole period is of critical importance.