What Is Split-Testing & How Does It Work?

Affiliate Disclosure: Look, I'm working my ass off delivering value to you. So, yes any links that you see in this post are probably affiliate links to help keep this blog going. They are true recommendations and if you click on them and purchase, I probably earn a commission (at no additional cost to you).

 

In online marketing, you often hear the word split-testing being thrown around.

And, it’s for a good reason, as split-testing can have great benefits on your conversions.

If your conversions are higher, then you’re going to make more money.

It’s that simple.

What is Split-testing?

What Is Split-Testing? 

Split-testing also is known as A/B testing is the process of testing different variations in marketing. 

The reason why some call it A/B testing is because you are testing variation A vs variation B to see which one generates more conversions. 

Before modern technology, you used to have to manually create both pages (or whatever you’re testing), drive traffic the same amount of traffic to them and then compare the analytics of each page.

However, today, technology is a little more advanced and most software will give you an option to test two different variations.

By doing this, the software will then split the traffic into an equal amount between the two variations and show you the analytics of variation a and b. 

Where Can You Split-Test 

You can split-test pretty much anywhere and anything you want, but here are some common places which marketers like to split-test. 

Emails

Firstly, we have emails.

If you’re building an email sequence, you probably want to know whether Email A or Email B converts best.

So, you can then split-test those emails.

For example, you could see if your story generates more clicks or Email B which is a more personalized email but doesn’t contain a story. 

ActiveCampaign and GetResponse even lets you split-test inside of your automation workflow which is great.  

Related Article: 

GetResponse vs ActiveCampaign

This can also work well throughout your entire automation, combined with lead tagging based on the automation variation they went through. 

If you split-test two different variations of the entire automation workflow and your leads with the variation b tag converted much better, then you know to use that variation of the workflow. 

I also like to test variations if I send an important email to see which one land in the inbox and which one land in the promotions tab. 

Then, I can use the one which lands in the inbox and get a much better open rate. 

Landing & Sales Pages

This one is super important.

On your landing pages & sales pages, conversions are probably the most important thing.

If you can change a sales page to convert 10% better, then that same sales page is generating 10% more money. 

So, by creating two landing or sales pages, then split-testing them, you’re likely to find which one does better. 

Two great platforms that let you split-test extremely well are Kartra and InstaPage. 

Ads

The next one is also extremely important.

It’s ads.

If you are paying for your ad to be clicked, you can drive the cost down by generating more clicks.

Google, Facebook, and the rest optimize their platforms to make money, so they sell the spaces cheaper to advertisers that generate the most clicks.

If your ad isn’t generating clicks and somebody else’s is, Google & Facebook are going to prefer that advertiser and reward him by lowering his cost per click.

When you are buying thousands of clicks, this is a huge advantage.

Plus, you can also split-test different ads to see which one and its audiences convert better with your offer or qualify the visitor better for your offer, then this also has a direct impact on your sales. 

Pop-Ups

Lastly, we have Pop-ups.

And, this is also a really important place to split-test.

If you have a pop-up on your blog which is converting at 2%, there might be a variation that converts at 4%. 

That’s a huge difference. 

Now, every 1,000 visitors, you are collecting 40 emails instead of 20. 

What Elements Can You Split-Test? 

You can split-test pretty much anything you want.

  • Images 
  • Videos 
  • Button Size
  • Button Font
  • Button color 
  • Font size
  • Font Color 
  • Background Image

Pretty much anything on your page or email.

However, what I usually like to split-test is the:

  • Headline
  • Offer
  • Discount
  • Video (If it’s very different or one page uses a video and one doesn’t) 

I like to test things that will have a bigger impact.

If your customer really wants/needs whatever you have to offer, then changing the button color and font size, will probably not make a big change in conversions.

And, unless you are testing with a huge data set of visitors, you won’t even know if that small change in conversions was due to whatever you were testing or just because the people were a little different.

This would then result in more testing and you probably wouldn’t know when to stop.

But, if you would split-test something which would make a bigger impact like the offer, headline, etc. then, you can find meaningful changes much quicker.

That’s just how I do it, I’m sure there will be some people who like to split-test those little elements too. 

Benefits of Split-Testing

Split-testing really just has one benefit which is that you can find what works. 

By split-testing you can find which variation:

  • converts best
  • generates the most amount of traffic
  • is opened or read the most

You get to experiment and see which one will work the best.

That is pretty much the only benefit, but it’s huge. 

If you can change a page or an email to even just convert 1-2% better in the long run, you’re going to make a lot more money. 

Final Thoughts

I hope this article has shed a little light on what split-testing is and how it works. 

It’s definitely a useful feature that is available in most sales funnel software and autoresponders which you should be using, once you generate a significant amount of traffic.

If you have an extremely small data set, I wouldn’t worry about it too much as you don’t know if the changes in conversions are due to your changes or if the visitors were just different. 

Leave a Comment