This is a guest post by Justin Clark of AdChop.com
Want a simple way to increase the profits from your website or online business? Split test! Not only can split testing measurably increase your profits, I’d go as far as to say that your online business will be stuck in neutral if you don’t do any split testing.
In this post I’m going to explain why and what to test, and hopefully get you started on how to test.
What Is Split Testing?
Split testing involves testing 2 or more versions of “something” (an ad, a headline, etc) to find out which performs best.
As an example, you might test 2 different headlines on your landing page (half of your visitors see one headline, and half see the other headline) to find out which headline generates the most sales. Once you discover which headline generates the most sales, you can start showing that headline to ALL of your visitors, and eliminate the poorly performing headline.
What Should I Test?
In a word: “Everything”
Of course you don’t need to limit yourself to testing only one aspect of a web page, and there’s no need to limit yourself to only 2 versions of whatever it is you’re testing.
Besides headlines, here are a few other examples of the sorts of things you can run split tests on:
- Different versions of PPC ads, to discover which ads get the highest click through rates (CTRs) and conversion rates (the percentage of clickers who become customers)
- Different email opt-in form designs, to discover which one gets the highest opt-in rate
- Different email headlines when you send an email out to your mailing list, to discover which headline results in the highest email open rate.
- Different Adsense positions on your website, to determine which positions have the highest CTRs.
If split testing sounds daunting, don’t worry – you can start simple and there are some free tools that you can use to get started (I’ll mention those tools later in this blog post).
Here’s a simple example of a Facebook ad split test from a real campaign that was run:
The Virtuous Circle Of Profit Growth – An Example
To demonstrate the value of testing, let’s look at a slightly more detailed example. This example involves a bit of math, but it’s simple and worth following along:
Let’s say your website sells an ebook for $20.
You’re doing some pay per click (PPC) advertising to bring traffic in to your website, and 1 out of every 100 visitors to your current landing page buy your ebook (in other words your conversion rate is 1%).
Now let’s say that through some testing you’re able to double your conversion rate so that now 2% of your visitors buy your product.
At first glance, it simply looks like you’ve doubled your revenue (and you have – that’s great!).
But there’s more to the story…
Let’s say that it costs you $15 in PPC advertising to bring 100 visitors to your website. With your original 1% conversion rate you’re getting one $20 sale for each $15 that you spend on PPC… so your profit for those 100 visitors is $5:
$20 Sales – $15 PPC = $5 profit
After you double your conversion rate from 1% to 2%, you’re now selling 2 ebooks (total sales $40) to the 100 visitors who are brought in for each $15 you spend on PPC:
$40 Sales – $15 PPC = $25 profit
So you see, although your conversion rate only doubled (from 1% to 2%), your profit increased 5-fold, from $5 to $25 per 100 visitors!
And there’s still more…
Your new higher profit margins can mean that:
- You can now afford to bid higher on your PPC advertising, perhaps outbidding your competitors and squeezing them out of the market, bringing in significantly more traffic to your own website.
- You may be able to afford advertising on new websites or ad networks that were previously out of reach, driving even more traffic and profits
- Affiliates (if you have any) will see their own profits rising and will be motivated to drive even more traffic your way, rather than to your competitors.
In this example we’ve only looked at testing and optimizing one step in your sales funnel – your landing page. Imagine what could happen if you also:
- Tested and optimized your PPC campaign so that you doubled the click through rate (CTR) of your ads, effectively cutting your cost per click in half.
- Tested and optimized your mailing list optin form and messages to increase the percentage of your buyers who become repeat customers
Combined, testing and optimizing all aspects of your sales funnel can have a significant effect on your bottom line.
What If Your Online Business Isn’t That Complicated?
What if you don’t have such a complicated sales funnel – what if all you have is a simple Adsense website, for example? Well then for you, your testing will be simpler, but still very worthwhile.
A couple things you might want to test are:
- The position, size, shape, and colors of the Adsense blocks on your website.
- Your page titles, to see how they affect your search engine ranking and also your click through rate from Google’s search results.
The next tip will help you no matter what type of testing you’re doing…
A Tip – Test Big, Dramatic Changes
The most worthwhile tests are generally those that compare the results generated by big, dramatic changes.
You’ll Get Bigger Improvements
Although there are exceptions to every rule, small changes generally result in small improvements, and big changes result in big improvements.
If you make a few small tweaks to a squeeze page you might increase the opt-in rate to your mailing list by 5%. If you test some major redesigns to your squeeze page, you might increase the opt-in rate by 50% or 200%.
To increase your chances of hitting a home run, test big changes!
You’ll Get Results Faster (And Cheaper)
To show you how big changes get results faster & cheaper, I’ll use an example from an advertising campaign:
Say you’re running 2 ads simultaneously to see which one gets the highest CTR.
There is only a small difference in the appearance of the two ads, and after each ad has received 10,000 impressions, Ad A has 50 clicks and Ad B has 55 clicks.
Which ad is better? The answer is, you don’t know – the numbers are too close to say for sure which ad will perform better if you keep them running. You’re going to need to run the ads for a lot longer (and therefore spend a lot more money running them!) before you have conclusive results.
Now instead assume that the two ads look radically different from each other. After 10,000 impressions, Ad A has 50 clicks and Ad B has 300 clicks.
Which ad is better? Because of the big difference, you can be very confident that Ad B will continue to perform better in the long run; there’s no need to spend more of your budget testing Ad A.
Examples Of Big, Dramatic Changes
Here are a few examples of what I mean by big, dramatic changes:
- Rather than testing whether you should put an exclamation mark at the end of your headline or not, try totally rewording the headline so that it conveys a very different message.
- In an online advertising campaign, rather than testing 10 ads that are all relatively similar to each other, try to make each of the 10 ads as different as possible from every other ad.
- Instead of moving an Adsense block an inch to the left of right, try moving it from the bottom of the page to the top of the page.
- Instead of making small tweaks to your written sales letter, test whether a written, audio, or video sales letter results in the highest conversion rate
Free Tools for Testing Your Website
Many ad networks (such as Google Adwords) and mailing list services (such as Aweber and GetResponse) have their own built-in tracking software that allows you to measure the results of your tests. If you’re already using those ad networks & mailing list services, you don’t need to pay anything extra to use their testing/tracking software.
Other free software that helps you run split tests & measure the results includes:
Google Website Optimizer – for testing & tracking the results of changes to your website
Tracking202 – for testing & tracking paid advertising campaigns (primarily for affiliate marketers)
What Should I Do Next?
If you’ve never done any split testing, start by picking one thing to split test that you think might possibly have a potentially significant effect. I suggest starting simple – don’t try to test everything right away because you can easily get bogged down. Once you’ve seen some positive results from your initial test, you’ll be more motivated to continue testing & optimizing other aspects of your online business.
If you’d like to see the results of split tests that others have run, check out the case studies on these websites:
AdChop.com – Case studies from ad campaign split tests
WhichTestWon.com – Case studies from landing page split tests
I hope I’ve convinced you that testing is worthwhile and have pointed you in the right direction to get started.
Justin Clark has been making money online for 5 years, specializing in pay-per-click advertising and affiliate marketing. To learn how to increase your website’s profits through split-testing, check out the free case studies on his site www.AdChop.com.