The Ultimate Affiliate Marketing Split-Testing Guide

Are you an affiliate marketer? Do you want to make more money? Yes and yes, right? You’ve probably tried every trick in the book, right?

Google image ads, Twitter traffic, cheap Facebook ads… every possible way of getting cheap clicks to your landing page has probably caught your eye at some point or another.

Most of these tricks don’t really work, and if they do, it isn’t for long.

The real secret to making big money as an affiliate is split-testing.

And as an affiliate marketer, it’s actually easier for you than it is for a lot of other people…

 

Affiliate Marketers Have So Much Less to Worry About!

Affiliate marketers are already several steps ahead of the online marketing curve.

That’s because, as an affiliate marketer, you aren’t responsible for the entire chain of conversion. All you have to do is find qualified leads, warm them up, and send them to the vendor’s landing page – the vendor will do the rest.

Not that all of that stuff is easy… that’s why some people are super-affiliates, and some people don’t make a dime. But still, it’s the vendor who has to create an awesome product, and to tweak their landing page and funnel until it is a high-performing conversion machine. Which means that you don’t have to worry about turning prospects into customers, or keeping them as repeat customers – that’s the vendor’s job.

Which is why things are easier for affiliate marketers – they don’t have to split-test their product ideas, and they don’t have to optimize the back-end of the funnel.

But despite all that, in some ways, split testing is actually harder for affiliate marketers than it is for other people…

 

Affiliate Marketers Have Much Less Control

Split testing is harder for affiliate marketers because there’s less that you can split test; here’s a short list of the things that you don’t control:
• The product
• The checkout process
• The back-end funnel
• The messaging
• The imagery
• And the list goes on…

So in other words, if the offer isn’t fundamentally capable of selling itself, there isn’t all that much you can do about it.

Which is why affiliates look for tried and tested offers, with solid and proven EPC (earnings per click) rates. 

The trouble is that every other affiliate marketer is doing the exact same thing!

And if every other affiliate marketer is doing the exact same thing…

 

There’s Dramatically More Competition!

Yup, competition is a lot tougher for affiliate marketers.

Whereas people who sell their own stuff have to compete with other products and offers (which is hard enough!), affiliate marketers also have to compete with lots of other people who:
• Target the same lead sources,
• Run similar ads,
• Refer to the same swipe material, and
• Sell the exact same thing!

Yikes – what can you do to stand out?

That’s where split-testing comes in. In order to excel as an affiliate marketer, you need to constantly be split-testing the things that you do control, namely:
• Your lead sources
• Your ads
• Your landing pages
• Your incentives

Let’s go through these all one-by-one, and explore what to test, and how to do it. But first…

 

How do you do split testing, anyway?

Just to get us all on the same page, let’s quickly review.

Split testing (or A/B testing) is about testing two different variations of something, to see which does a better job of getting your audience to do what you want them to do. Common examples of things that you can split test include headline variations, button text and colors, and product pictures.

At a high level, that’s all there is to it.

Now, when it comes to running the actual tests, you need two things:
1. Some sort of technology to randomly serve the different variations to visitors.
2. Some way of analyzing the results to know when they’re statistically significant (in other words, to make sure that the results you’re seeing are caused by whatever you’re testing, rather than just being a random coincidence).

This may sound a bit complicated, but you really don’t need to worry about it; there are great tools that will do it all for you, like Google Website Optimizer (free), or Optimizely and Visual Website Optimizer (paid, but are worth it).

And as for doing the analysis, there are lots of great free split test checkers out there that can do it for you.

Okay, are we up to speed?

Good – now let’s get into the meat of what you, as an affiliate marketer, can be testing…

 

Split Testing Your Lead Sources

The first thing that you can (and should) split test is your lead sources – in other words, where are you getting your traffic and prospects from.

Too many affiliates are one-trick ponies in this regard – they have their favorite traffic strategy (often the only one that they’ve been able to get decent results from), and they stick to it like glue.

Well, there are a lot of traffic sources out there, for example:
• Google AdWords
• MSN/Bing/Yahoo!
• Facebook
• Twitter lists
• CPV ad networks
• Email blasts to your list
• Email blasts to rented lists
• JVs with related products
• And the list goes on… and on… and on!

Different lead sources will work differently for different offers, so test them to see what works best. Also, when testing, make sure that you aren’t comparing absolute conversion numbers, but rather conversion relative to the cost of the traffic!

 

Split Testing Your Ads

The second thing that you should split test is the ad or call to action that you are presenting to your leads.

Here are some things that you can test:
• The headline of the ad. Try featuring different pain points or benefits, to see what works best.
• The body of the ad. It may be short (like on AdWords) or longer, but either way, you’re communicating what you’re offering, and why it’s valuable. Change that up to see what makes a difference.
• The call to action. Experiment with more and less explicit calls to action (“click here” vs. “learn more”, for example), and experiment with a focus on action versus results (“buy now” vs. “get instant access”).

Just one thing – when you’re testing different ads, make sure you’re tracking the right thing, and that is sales (not click-throughs)!

It’s not hard to write an ad that will get lots of people to click, but what you want to do is write an ad that will get few people to click, many of whom will then buy. Consider adding barriers to interest (like a listed price) on the ad, to deter people who aren’t likely to make a purchase.

 

Split Testing Your Landing Pages

Yup, direct-linking to an affiliate offer is a sucker’s bet, because you’re giving up way too much control over the sales process.

Which is why your landing page is where most of your split testing is going to happen – and so it should be! Your landing page, realistically speaking, is what will make the most difference to your bottom line.

Here are some of the most important things that you should test on your landing page:

• Your headline. Yes, you’ve heard this before, but it’s true – the headline determines how many of the people visiting your page actually read the first paragraph, as opposed to clicking away. So test, and test, and test, until you get it right.

• Your page format. In other words, short vs. long copy, and text vs. video.

• Delayed button introduction. You’ve probably seen this – a button that only appears after a video has been running for a certain amount of time. Test it out – it might work for you.

• Your colors. Different colors prompt different behaviors (this is called color psychology), and you should test different combinations to see what works best for your offer and audience.

• Your button placements and text. This can make a huge difference on click-throughs and purchases, so test, test, and test!

• Your trust seals. These are the seals from Trust-E, Verisign, the Better Business Bureau, and other organizations, that let the reader know that they can trust you and the merchant that you are representing.

• Your affiliate disclaimer. This is one that many people overlook, but it can make a big difference. Sometimes you’ll get better results with the small print in the footer-style, but sometimes you’re better off working it into your text (“Hell yeah, I’m an affiliate – I’m proud to stand with this offer!”).

This can take a lot of time to do properly, but believe me, the results are worth it!

 

Split Testing Your Incentives

The last thing that you should be split testing is something that many affiliates overlook altogether, and that is your incentive offer.

An incentive is something that you offer so that people buy through your affiliate link, as opposed to going directly to the merchant, or going through somebody else’s link. This is particularly valuable during a launch, when your prospects are likely to be hearing about the offer from a lot of other people.

Here are some of the things that you could consider offering as an incentive to people who buy through your link:
• Free or discounted access to one of your products
• Some free consulting time with you
• Access to an exclusive webinar that you will put on
• A portion of your affiliate commission, thereby reducing the effective price of the offer
• There are many other ideas… use your imagination!

The importance of an incentive can’t be over-stated – especially in an affiliate-saturated market, it can make all the difference.

Now, after reading this post, you should be ready to go out and split test to your heart’s content. But before you do that, I have two warnings to leave you with…

 

Warning #1: Testing Shouldn’t Be Random…

Some people think of split testing as being akin to panning for gold – you grab a big chunk of dirt, and hope there’s something valuable in there.

So they test every random thing without any rhyme or reason, until they arrive at a combination better than what they’re already running.

This might work if you’ve got the traffic of a Google or Yahoo!, but I’m assuming that you don’t (and even if you did, it isn’t the smartest way to do things).

Your split testing should be planned and intentional; think about what is likely to have the most impact on your audience, and there for yield the best results.

If you’re not sure where to start, then use a framework to track your audience through the conversion process, to see where you need to start first – or ask someone you trust to weigh in and give you some advice.

 

Warning #2: Only Test ONE Thing at a Time!

After reading a post like this one, you might feel the urge to make a long list of things you want to test, and then create a new campaign implementing all of those tests.

Resist that urge!

The key to split testing is to isolate variables – in other words, you want to test one thing at a time, so that you know what is responsible for any changes that you observe.

Order your list by priority of what you think will make the most difference, and then test one thing at a time, until you are confident that the results are statistically significant (as described above – you can use free split test checkers to make sure that this has been accomplished).

So plan out your tests. Rank them by priority.

And start testing!

Dedicated by: Danny Iny

About the Author

Danny Iny is an author, strategist, serial entrepreneur, and proud co-founder of Firepole Marketing, the training program that turns non-marketers into expert marketers. Visit his site today to download a free split test checker, or follow him on Twitter @DannyIny.