How Split Testing Can Double Your List Growth
Steve Shaw Tuesday, November 27, 2018
Share this page with a friend
The simple truth is, if you’re not split testing your opt-in offers, you’re more than likely leaving at least half your potential subscribers on the table. For many businesses, that equates to something like half your revenue.
Don’t believe me?
Brian Dean ran some split tests on popups he was using for lead capture, as a result doubling his conversions.
Glen Allsopp did the same, testing a simple headline change for an opt-in offer, and his income has since increased dramatically.
Countless others are doing similar all the time.
To be clear, it means you’re getting double the number of subscribers from the same amount of traffic. It’s not hard to do, takes just minutes, and the results can be profound.
In fact, doubling your list growth is really just the start.
For example, I’ve been running some tests on one of my blogs. It gave me an increase of 391%—so nearly 4x as many leads.
But I didn’t stop there. Some further tests have now doubled that. In other words, 8x as many leads as the blog was attracting before the testing.
So, if you’re not doing so already, it’s time to get serious about split testing.
Reasons Why You’re Maybe Not Be Split Testing
People are sometimes afraid of starting to split test, fearing it will take a long time to set up, that it’s overly complex, or that they’re unlikely to get much benefit.
None of these are true.
Split testing your subscriber offer can literally take just a few seconds. Then you simply keep an eye on the data as it comes in.
It can get pretty exciting to watch. As soon as you’ve ran a successful test, you’re far more motivated to run more—you know it works!
It’s not complex either. As this post shows you, it can take two or three clicks to set up and start a test running. Depending on the test you want to run, it may take no more than ten seconds.
And if you think, despite all the examples of successful case studies all over the web, that somehow you’ll be the exception and not get much benefit … maybe it’s time to think again.
Reasons Why You Should Definitely Start Split Testing
A lot of business owners and marketers approach growing their business—and growing their list—by solely focusing on getting more traffic.
Of course, that’s worth continually working on too. But, compared to split testing, it takes longer and can be a lot more expensive.
For example, to increase traffic, you might be creating lots of content and/or paying for ads through different channels. And even though that might give you more traffic, it’s not necessarily translating into corresponding growth of subscribers on your list.
On the other hand, split testing allows you to potentially double or more your subscriber conversions—the number of visitors you convert into subscribers—without increasing your traffic at all.
And the return on your investment in growing your traffic will likely double too.
Let’s say you’re paying for 100 visitors at $1 each, and the page converts visitors into new subscribers at 20%. In other words, you’re getting 20 subscribers, with each subscriber costing you $5.
With split testing, by doubling your conversions, you’re now getting 40 subscribers costing you just $2.50. That makes a huge difference to your business’s cash flow and your ability to spend on new advertising or other ways to get more traffic.
So split testing not only grows your list—but can have a big impact on your ability to grow traffic too.
Doing the opposite—i.e. not split testing—means you’re harming not only your list growth but also your potential traffic levels, even if traffic is what you think you should be concentrating on instead. You need to do both.
Even better, the more traffic you get, the easier and quicker it is to split test. With more data to play with (i.e. more visitors and more visitor actions), you can reach meaningful conclusions more quickly.
Finally, let’s face it. If you’re not testing, you’re basically flying blind. You’re assuming that your own judgement is superior to actual scientific data that can prove one way or another.
Obviously, that’s not the most effective approach to growing your business.
How hard is it to double your conversions?
You might be fortunate enough to run a test that doubles your results straight away. For example, Glen Allsop’s winning headline converted at 54.2%, compared to what he had previously, converting at 24.2%. A more than 100% increase.
But not all tests give you these kinds of results.
You might get more modest increases of say 20-30%.
But this is also where the magic happens.
Through compounding results, you only have to run three successful tests, giving you increases of 20%, 30% and 30%, to get a 100% increase overall.
This is why it’s so important to split test continuously. Even when individual results are more modest, they compound into something far more meaningful.
Of course, not all changes you make will beat what you have already. You need to expect that to happen, and not let it put you off further testing. After all, it’s good news! It simply confirms that what you have right now is the best version—until the next test where you may get a 50% increase.
You just never know what might happen until you test! The results might knock you off your feet and can be very exciting.
How to Split Test
Firstly, decide what you’re going to test. The headline’s a good place to start if testing say an opt-in offer, but lots of other changes can have a meaningful impact too.
- Where and how a dynamic opt-in offer appears—for example, you might want to test:
- A lightbox popup versus a floating bar type offer on your website.
- A lightbox popup appearing on entrance versus when the visitor’s about to leave. Or even immediately on entrance versus say 5 seconds after arrival.
- Floating bar appearing at the top versus the bottom.
- The layout of a landing page and the different elements it contains—for example, adding a testimonial or two can have a big impact on conversion rates. If you already have testimonials, test using a different one.
- The color of a form button and the text it displays.
- The number of fields in your opt-in form.
- The display of a privacy statement or link, the wording used, and any icons such as a padlock.
- Images used.
And so much more. There’s always something else to test, and it’s always surprising how small changes can result in significant improvements.
In terms of how to test, most list building tools and lead capture systems have split testing facilities built right in, and you can simply follow their help information for setting up tests. Here are just a few of those available that provide split testing abilities:
4 Top Principles of Successful Split Testing
Test One Change at a Time
Generally speaking, only test one change at a time. Otherwise, the benefits of one change might be cancelled out by the negative impact of another.
You could end up with a result that shows little or no change, whereas testing just one of those changes could have given you a lift of 20%.
Don’t Set Up Multiple Variations
Don’t set up more than one variation of your test. Along with the control (what you have now), you should have just one other version you’re testing against.
If you have more than one version to test against, it takes a lot longer to get meaningful results from your data. That’s fine for the likes of Facebook and Amazon because of the sheer volume of traffic they attract, but not for most of us.
Wait for Statistical Significance
Only stop running the split test when you know, with scientific certainty, that your results are valid. That means you should never stop a test early, simply because one version appears to be the runaway success.
More often than not, this type of result appears early in the test, and then results begin to stabilize and can even tip the other way completely by the end. I’ve seen this happen multiple times in my own testing.
It means that by stopping early, you’re potentially reducing your conversions, not increasing them.
So you have to wait until you have what’s called statistical significance—in brief, that means you have scientific proof that you have a winner.
The Right Mindset
Finally, maintaining the right mindset is crucial for successful split testing. If you let a couple of mediocre or negative results put you off, you’ll never achieve the profound and lasting benefits that split testing over the long term can give you.
The truth is that all tests tell you something meaningful about your market that you can then use to improve your conversions. Just try something different and run another test.
More Articles By Steve Shaw