Split testing is one of the last things that you need now, right?
Because from what you’ve seen so far, you’ve concluded that it’s a total waste of time.
But, I bet there was once a time when you began to run out of meaningful tests to run, which is actually the reason why you are currently seeing little to no gains at all.
Multivariate testing, popularly known as “split testing”, is an excellent way to see what marketing campaign in your business will boost conversion rates.
Remember when Albert Einstein reportedly stated that, “the most powerful force in the universe is compound interest?”
Constant testing of your marketing efforts and campaigns leverages that force.
Now, if you could get just a 3 percent enhancement per week from your social media, PPC, or email campaigns, what do you think that would result to at the end of a year?
Well, assuming you started out in January earning $500/campaign (maybe with an email campaign), a 3 percent improvement per week would have your last campaign of December earning you about $2325.50. As you can see, that’s over a fourfold increase in sales from a very little uptick repeated every week.
So, who wouldn’t want to take a shot at that?
Fortunately, to help you solve this issue, I’ve assembled 17 different split testing ideas that you can start implementing right now.
These split tests will give you real, actionable results. And, to make it super easy for you, I’ve broken the 17 split testing ideas down into 6 categories:
- Email marketing
- Load time
- Landing pages
- Social media
- Copywriting and content
So, depending on what area of your business you need testing, you can just scroll to the relevant heading.
Call-To-Action Split testing ideas for increased conversions:
1. Call-To-Action Button
One of the most important and under-tested areas of a website is the “call-to-action” section.
Lots of buttons are randomly tagged something like “Learn More” most of the time. But why not experiment a little with more personal options to find out if it will eventually boost conversions?
There are a lot of tweaks you can give your CTA text. Learn how to make it more personal instead of the regular “Learn More”, “Download”, “Sign Up”, “Buy Now” ones.
You could try something like “Yes, Give Me My Freebie” or “I Love Free Sutff” and see what difference it’ll make.
Apart from what’s written on your Call-to-Action, another critical thing you always need to test is the button color.
You’ve may have seen this split test performed some years ago where a red button beat a green one in terms of conversions, keeping the tester astonished.
However, call-to-action button color goes deeper than that.
For instance, assuming you have two call-to-action buttons – a “free trial” and a “buy” button. Your ultimate goal is for people to buy your product, although, you wouldn’t be disappointed if they ended up downloading a trial first. Yet, you want to enhance the buy option to make it more sexy and enchanting.
What should you do?
In a situation like this, the “Download Now” option is considered less important, even though it’s the first call-to-action on the page. Instead, you want the eyes of your visitors to be attracted to the more prominent “Buy Now” button.
Try muting the color of the button you want to give less emphasis to and see the difference it’ll make in your conversion rates.
2. Call-To-Action Button Placement
When it comes to creating an effective call-to-action button, positioning matters a lot. Many people often wedge their call-to-action buttons in between blog posts, slideshows, and headlines to the extent that it can be highly overwhelming, especially for first-time visitors.
However, split testing your button placement will give it some breathing room, similar to how PayPal does with their calls-to-action:
Proper use of empty space lets you focus your eyes and attention more clearly on the colorful CTA button more than a page filled with confusing information.
Implement this into your own landing page, play with different button placements and see what happens.
3. Directional Arrows
Notice in the first example above that the call-to-action button has small visual and navigational cues instead of having just a plain button?
That small visual improvement serves a purpose. Directional arrows (like “down” for download) make it very clear at a first glance what the intended action should be.
Most times, call-to-action buttons that want you to take a different action beyond downloading will have a right-facing directional arrow, signifying forward movement.
This is an example from InstaPage:
This idea is worth trying.
Who knows? It could be the thing that will skyrocket your conversion rate.
4. Plain Text vs. HTML
This is one of the oldest debates of the 21st century – which of these two gets more response?
To me, it totally depends.
For instance, your email marketing service provider may let you design a page with both HTML and plain text elements, but it may equally insert its own proprietary code.
In turn, this can result in having rendering problems in both software and browser-based email programs which can be problematic for an email campaign.
Litmus created an excellent infographic that demonstrates the processes an email must undergo and all of the flaming circles your email message must jump through before it’ll be received.
Things like mobile email devices and spam filters see HTML emails as potential problems, and they filter them accordingly.
At the same time, it’s not advisable to just copy and paste from your word document program, as a lot of those programs often insert their own code that makes no sense to email or browser software. This is yet another potential issue for an email campaign.
Instead, the best approach is to test your email marketing messages by offering the user a choice by presenting both the plain text version and a web-enhanced version of the email message.
5. Dynamic Content Emails
People respond differently to personalization. Some view it as a marketing ploy, while others see it as an impressive change from the usual “Dear valued customer” messages they get.
For example, you could A/B split test the following in your email messages:
- Using no name at all versus the subscriber’s first name.
- Adding the company name in the “From” section versus using an employee’s name who’s in charge of customer relations.
You can also test sending out messages to fans of your Facebook or Twitter pages and mention your gratitude for following you.
Try to see where personalization takes you.
6. Email Calls-To-Action
Your email calls-to-action should also be given proper attention. For instance, does a text link perform better than an image-based call-to-action?
What about having both in your email messages (for those who turned off images in their email)?
Where do people click in your email message?
Test the text or button placed above the fold and at the bottom of the message. The open rates of your email might see an increased conversion rate just by playing with this simple stuff.
Remember that you won’t know exactly what works best until you try random variations.
Website Loading Time
7. Tag Management
Tag management is one of the things many people plan to “work on” when they have the time. But, getting to it right now can help you to improve your website loading speed and encourage better split testing results to boost conversion rates.
And it’s pretty easy to do.
It requires a little technical know-how, so it’s advised you contact your web developers and have them check it out. They’ll have to check out all of the instances being called in your “<head>” tag before you dive right in.
But, the final result will be:
Split test a version with your current setup versus a single container to see how much time you could be shaving off your server load. If it’s a significant amount, make it permanent.
Check out the link below to learn how best to implement this:
8. Determine What Needs To Go
It can be difficult to decide what’s worth cutting out on your pages and what’s worth keeping.
A recent study confirmed that the majority of website users expect a site to load fully in 2 seconds or less. Consequently, they abandon websites that load in 3 seconds or more.
However, you can use some in-built caching plugins and compression algorithms to reduce stress on the server, thereby improving your site’s load time.
Edit everything ruthlessly.
Test removing anything that isn’t 100% relevant to getting the user to act or click. That can even mean things like getting rid of that beautiful slideshow on your homepage or even your navigation menu.
Interestingly, tablet users are in the center of this “less is more” fight, so what about them?
The guys at Shirtinator decided to redirect the tablet users of their site to their website’s desktop version that’s optimized with HTML5.
So what happened?
You’d be wrong if you guessed that conversions went down.
In fact, this version beat mobile for tablet users by more than 70%, raking in an extra 32% in the number of completed orders.
Sometimes you don’t need a few little changes to increase a website’s conversion rate, but rather a radical modification in the way you approach the design of the page itself.
For example, Marian University continuously tests different versions of its landing pages to some astounding results:
Don’t be scared to experiment with your landing pages. Start to think outside the template.
Most times, the biggest changes bring the best results.
The Best Tools That Will Help You Build A Profitable, High-Converting Landing Page In Minutes
10. Geo-Targeted vs. Generic
This idea was originally made famous by local daily deal sites such as Groupon. Nowadays, geo-targeted content has taken on a life of its own to display almost everything, from apparel to car parts.
However, geo-targeting isn’t limited to physical location alone; things like language and currency can also be seen as geo-targeting.
So, will your prospective users be more captivated by your website if you find out where they’re from and mention it on your landing pages?
You can only find out by testing.
11. Copy vs. Video
The team at Mindvalley Insights recently wrote a very thorough guide about using a regular sales copy in landing pages versus video.
From their studies, video performs better than copy every time. However, there are still some instances where copy outperformed video.
Below is an excerpt:
But even if you choose to use video, it’s advisable to test the type of video that you use.
For instance, will you use a more personal one-on-one style or a whiteboard-style “explainer” video like the one Rand from Moz uses on his “Whiteboard Fridays“?
You never can tell which will work better for you until you test them.
12. Trust + Social Proof
Can logos like the usual “hacking-safe” or “transaction-safe logos” increase your landing page’s conversion rate, or will they distract your prospects instead?
Some studies claim they help, while others say they don’t impact the bottom line at all.
However, you should test adding a few recognizable and reputable badges on your landing page and see what your own users think.
Similarly, you have to test and see if having some sound social proof on your landing page can help boost your CTR and conversions. A study by Consumerist confirmed that “nearly 70% of consumers rely on online reviews before making a purchase.”
Therefore, putting a few social proofs like customer testimonials or media mentions on your landing page could greatly improve your conversions.
You’ll see a good example of this when you visit the FreshBooks website:
13. Form Images
Almost every landing page contains a call-to-action form for a demo, free download, free trial, or other offer.
But, it’s good to split test the inclusion of an image with your form. It could be a photo of you or a user using the product, a photo of the product itself, or even a landscape shot citing your product.
You can also just use a cool image that is highly relevant to whatever product or service you’re promoting. The idea is to ensure the image is very attractive and reflects your product.
Check out this one used on the Match.com homepage:
Who said landing pages are just boring forms?
Copywriting and Content
The most vital aspect of your split test campaign is probably not in the slick CTAs or the flashy graphics, but in the words that you write.
Here’s the fact: People don’t have the time to read every single word on the internet. But that’s not really an excuse to be sloppy with your writing.
Use these few split testing ideas to analyze your site and see how it performs.
14. Copy Length
This has been the rallying cry of blog post creators, copywriters, and marketers for ages.
Which copy length sells better?
From my experience… it depends.
It depends not just on who you’re trying to engage, but also on your chosen market.
Some time ago, the guys at Marketing Experiments created a “long versus short” copy matrix which used empirical data and different case studies from around the web to reach their conclusion:
In summary, if you’re selling really helpful and “want-oriented” products that require less commitment and investment, you’ll perform better with short copy.
While “need-oriented” points, particularly housing, medical, and insurance-related offers do well with longer copy.
Not sure which segment your market belongs to? Split test using both long and short copy and see which outperforms the other.
15. Free Sample Offers
We all know how tempting free sample offers can be, and this form of marketing has enjoyed even more popularity on the internet in recent times. But, how can you accurately split test this without lessening quality?
Try split testing your sample together with an offer.
For instance, assuming you’re selling a skin care supplement and you are giving out some free samples of a hair care cream. Why not split test your offer in such a way that instead of giving away the hair care cream for free, you offer your users one month’s worth of the supplements for just $1?
Most times, people are more happy to pay for an offer that is very closely aligned with their needs than getting a freebie that isn’t.
16. Broad vs. Segmented Offer
If you’ve taken the time to study paths-to-purchase, sales funnels, and other online behaviors, you’ll notice that broad offers are not always the best choice, especially if a customer returns to a website or spends an unreasonable amount of time searching for something.
In cases like this, segmenting your offer to perfectly appeal to that particular customer may be worth split testing.
A good example of a company that did this on their support pages is Dell.
On their first version of the page, visitors saw the default support page with ordinary topics and FAQs.
On the second version, the page was triggered automatically and contained a live chat option if the customer had spent a specific length of time on the support page or had visited other topics within the support page.
Dell’s idea for this segmented split test was that if the customer couldn’t find what they were searching for by searching support articles or browsing the FAQs section, they’d be more likely to start a live chat session with a technician to assist them.
Having different options based on user’s behavior is a brilliant split testing tactic that can not only help you make the sale and earn that sought-after conversion, but can equally give you some priceless details about your potential customers.
17. Personal Experience/Review
We’ve discussed a lot about the possible copy tests you can run on your own website. But, what if you’re promoting another person’s product as an affiliate?
Split test the option of sharing your own personal experience with the product:
- What did you get from it?
- Were your expectations met?
- Why or why not?
If you’re promoting another person’s product/service, you will definitely see different reviews about it all over the internet. So, if you haven’t tested the product yourself, you could just leverage the reviews of other people (but don’t steal!).
However, if the product in question is yours, you can simply ask for reviews from your happy customers, and then put them on your website or landing page just as the guys at MarkHor did on their website:
You never can tell. Such a simple tweak to your copy could mean a lot for your conversion rates.
Split Testing Ideas
Now you don’t have any excuse for not knowing what to split test.
You can put any one (or all) of these 17 different split testing ideas to work on your website immediately and figure out which your audience prefers.
When you find one that works, make it the new baseline and split test some more ideas to see if you can you crank up conversions even more.
What are some of your favorite split testing ideas? What else can you recommend trying? Let’s hear your thoughts in the comments below!
Like this post? Don’t forget to share it!
Here are a few hand-picked guide for you to read next: