Foreword by Matt Diggity:
Last month at my company’s retreat to Bali, we set a pretty sizable goal for ourselves. That goal was to double our company’s affiliate revenue in three months. Two weeks later, we had already gotten half way there. We were on track to meet our target two months ahead of schedule.
How did we do it?
Rankings and traffic for our money sites were relatively flat during that time. Instead we decided to go all in on conversion rate optimization (CRO) and my business partner Kurt Phillip went straight-up beast mode. In two weeks’ time he was able to increase revenue 50% on our affiliate sites by testing the living hell out of our traffic.
Meet Kurt. Kurt has over 10 years’ experience in A/B testing and is one of the most talented CROs I’ve ever met in the field. Conversion optimization is so often overlooked in SEO, I felt compelled to urge Kurt into writing this guest post.
Without further ado…
Introduction: Kurt Phillip
Being the head of CRO (conversion rate optimization) for LeadSpring, I see, on a regular basis, how much benefit even basic CRO has for a website’s conversions and profitability.
CRO, also known as A/B or split testing, is one of the most neglected methods of increasing income from SEO. Once you really understand user-behavior and you start to develop a framework behind CRO, you’ll see it goes hand-in-hand with your whole SEO strategy.
Step 1 is getting people to your site via SEO. Step 2 is getting the users to do what you want them to do, such as buying something or subscribing to your list.
It really surprises me how much time people put into setting up a nice looking website, writing great content and getting it to rank on Google, yet they neglect to spend the same time on CRO.
Let’s say your website is making $10,000/month. If you increase conversions by 20% (over a few months), that’s an extra $2,000 per month or $24k over the year. You can buy a car with that. Heck, that’s most people’s 9-5 income after tax.
That is the power of CRO.
20% is not a blown out of the water increase either. Time-and-time again I see 100-200% increases in conversion rate.
It’s a shame, but most people I’ve talked to have said that CRO seems “too complicated” to figure out. It may seem like a daunting task to get started but it isn’t at all. You just need a framework and a process. You need a system behind what you do that it is repeatable, scalable, and most of all, simple.
The First Step: Assume You Know Nothing about Your Visitors
You know how to build WordPress sites, great!
You know how to build niche websites and provide unique value to the market place about a certain topic, even better!
You know how to implement an SEO strategy that sends your website to the top of Google so that eyes are on your website (and your affiliate links). Rank and bank, baby!
The next question is: Why did you decide to make your call to action button look like this?
Is that the best possible button to put in front of your visitors?
Even though I had been a web designer for many years, I was making decisions like this based on habit and what I’d been taught.
Is it possible that “send me some awesome info” might convert better than “send” or “submit”? Who knows?
I know. Because I use split testing to determine what actually works on my websites.
The great thing about split testing is that you get the users of the website to show you what they think. You don’t ask them hypothetically, you watch what they ACTUALLY DO. With the results, you then make adjustments to your website that allow you to convert, 100%, 200%, or 300% better.
Those aren’t inflated internet marketer numbers. I see increases like this all the time.
On one of our affiliate websites we have four identical call-to-actions (CTA’s) running down the length of the page.
The original call-to-action looked like this:
>> CLICK HERE TO VISIT OFFICIAL SITE <<
Split Tested Version:
As you can see, the split tested button was much more effective. Each placement of the CTA showed an increase in conversions ranging from 85.8%-239.9%.
Benefits of CRO
- More conversions for your website
- More sales
- More subscribers
But for me, the greatest advantage of CRO is that once you run an experiment and you have a winning conclusion, the increase in conversions kicks-in right away. Not in 3 months, today.
It can take months to see the monetary benefits from SEO. Sometimes you don’t see the full potential for up to a year, but with CRO, you see the increase in cash-flow straight away.
“If you increase your conversion rate, the benefit kicks in today. Not next month. Not next year. Today.”
Getting Started with Split Testing
Split Testing Fundamentals
Split testing is a method of comparing two versions of a webpage against each other to see which one performs better for a desired outcome. This could mean directing a user to an affiliate offer sales page or a purchase on an ecommerce store. Traffic is split between the two versions of the webpage and results are measured.
One of the greatest aspects of split testing is that you use data of actual user behavior to make decisions. It takes your own assumptions and biases out of your design decisions.
If you put your ego aside, you’ll often see how wrong you can be when you split test elements that your intuition was dead positive about. Uncovering weaknesses like these can make a huge difference to the profitability of your website.
Only Test One Thing at a Time
The key here is just to keep it simple… really simple.
“When testing, just change one thing at a time per page.”
The hardest thing about CRO is staying disciplined about testing only one variable. If you change too much at once, you won’t know which variation caused your result.
When you first start testing I’d recommend starting with a control (the way your site looks now) and then make one simple variation of it.
This may seem a little complicated, but when using Optimizely it’s as simple as drag-and-drop while their algorithm does the rest.
Use Heat Maps to Figure out what to Test First
Heatmaps is a smart little plugin that allows you to visually see where the most click activity occurs on a page. It will highlight your screen, using various colors which depict different amounts of interaction. You can then see which areas of your pages have the highest activity and will therefore be the first areas to run split testing campaigns on.
Download the free wordpress plugin called Heatmaps and create a free account.
When you have installed the Heatmap plugin and activated it, view your actual website while staying logged in and click the “heatmap” icon at the top of your screen.
Toggle the “Record” button to begin tracking clicks. You will have to do this for each page you want to record. The free version allows you to record up to 5 pages.
Depending on your traffic, in a few days’ time, you will know where the most frequent interaction locations are on your site, and you can then make data-based decisions on what to test first.
Setting the Foundation First
There are a number of standard conversion must-have’s that should be added to your site before starting your campaigns. These site features require no tests to determine if they should be added. Time has shown us that they just plain work.
They are, in no particular order:
- Clear above-the-fold headlines and images describing what the site is actually about. A visitor should know, within a split second, that they’re in the right place when they visit your site.
- Clear call-to-actions for each page: big visible buttons or bolded large contextual links in the body of the text.
>> CLICK HERE TO READ MY REVIEW OF CRO FOR SEOS <<
- No huge “walls” of text with no call-to-actions or images to break it up.
- Media throughout the entire page, such as images and YouTube videos. People rarely read anymore. Give them something to look at.
- A custom logo
- Favicon (people add sites to their bookmark bars and revisit to buy later on).
- Credibility indicators and recognition badges – “Safe and secure”, “Quick shipping”, “trusted seller”.
A good idea is to look around at your competition and mimic them as a baseline. Most of the time, your competition, especially if they are at #1, have split tested their site already.
Setting everything up in Optimizely
Optimizely makes the formerly complicated process of A/B testing so simple, anyone can do it.
Go over and make a free account, and then follow the steps in the video below to make your first experiment.
Again, please remember to split test only one element at a time. If you change too many variables at once, there will be no way to interpret which changes caused your results.
A classic example is to test two buttons’ texts against each other.
To change the text on the button, simply highlight the text, edit HTML and change it to what you want. It’s as simple as that.
Despite this being such an easy thing to do, remember, it is not uncommon to see 50-150% increases from just changing the text on a button.
Stacking Test Results to Find the Perfect CTA
I will outline a really basic example of how to split test call-to-actions on a page.
Let’s say you’re optimizing a review page for an affiliate product.
Test #1: Text vs. Button
>> Click to visit the official page <<
Now wait for Optimizely to conclude the experiment and determine a winner.
Awesome, variation #1 killed it!
Now that you have a winner from those two, update your site with the new button your customers really seem to like.
Test #2: “Best” Price or “Lowest” Price?
Now you’re going to change the text on the button. Remember to only change one thing at a time so you can isolate what change caused the conversion increase.
Are people looking for the lowest price or the best price? Who knows? Test it.
The variation didn’t work like you thought it would, so stick with the original “click here for lowest price.”
Test #3: Color
Now you test out color. Should you use a color that blends in with your site theme, or a contrasting color that pops out and draws the eye?
Let’s test it.
Red wins the day.
I’m sure you get the point by now.
Get creative with it. Before you start a testing campaign, come up with a long list of tests to run so you can keep rolling out continuous experiments over time.
As you continue to conclude experiments, your conversions will start to skyrocket and compound due to the constant improvement over time.
A List of Elements You Can Start Split Testing Today
The multitude of factors that can be split tested is virtually limitless. Below you’ll find a list that should get some ideas churning for your upcoming CRO campaigns. Some of the plugins listed even have split testing features built into them.
Other things to Test
- Modal Popup Window with call-to-actions that pop up after a certain amount of time
- Modal Popup Window that pops up as the mouse cursor heads towards the address bar or backspace button alerting the plugin that the user is about to leave.
- Sticky Widgets Plugin that makes sidebar widgets that follow the user as they scroll down the page. Keeps the call-to-action in the visitors face.
- Hello Bar Plugin that adds a bar to the top of the screen with a call to action.
- A well-designed version of a page vs. an ugly, outdated-style looking page. The results will often surprise.
- Having an “About Me” widget in the side-bar, split tested with and without an image of the “author”.
- “Buy Now” vs. “Add to Cart” vs. “Purchase now”
- “Official Site” vs. “Visit Now” vs. “I’d Like to Visit the Official Site”
- Countdown timers for an offer to push a sense of urgency.
- Adding a discount coupon on the site.
- Multiple call-to-actions as opposed to one.
- Buttons vs. textual hyperlinks
- Test colors, shapes and sizes of buttons. Don’t know Photoshop? Here is a great tool: Button Optimizer.
Always remember that no idea is right or wrong. Just test it and the users will tell you what they like best.
The Proper Mindset of a CRO Expert
Always wait until the Experiment is Complete
You wake up one morning, full of energy. Open your Optimizely account to see that your test is starting to yield results. The blue button is out-performing the red button by 68%. Considering that you make $100 per day on this site, this would mean that if you changed the button to be blue, you could potentially earn $168 today. If we look at that over the month…
Hold your horses, son. Let me introduce to you a little buzzword called statistical variance.
Statistical variance gives a measure of how the data distributes itself about the mean or expected value. Unlike range that only looks at the extremes, the variance looks at all the data points and then determines their distribution. Credit: explorable.com
In short: different days of the week bring different conversion rates. In fact, different times of the day yield a different consumer psychology and thus yield a different result. We need to gather enough data such that Optimizely can call a winner with a reliable probability.
Trust that Optimizely will do its thing and let you know when the experiments are concluded. It is all built into their algorithm.
How to Know When a Split Test has Concluded
Tests should be run long enough to gather enough data to confirm the significance of the result.
You’ll know when the test is complete because inside Optimizely, the result box will either turn green for a winning variation or orange for a loser.
Set Goals and Take Action
I really enjoy setting up campaigns and running disciplined tests as it removes all emotion from the decision making process.
To get the most out of your CRO tests you need to have a plan. You need to ask yourself, “What are the goals for my website?”
If your goal is to make $10,000/month in your niche and your site is currently making $1000/month in position #5 converting at 2%, getting to your goal becomes a simple numbers game.
For example, you can easily hit these numbers by increasing conversions to 5% and move the website to #1. Position #1 will bring in roughly 32x more traffic while a 3% increase in conversion will put you well past your milestone.
Matt Diggity and I setup this exact goal for one of our Leadspring projects and blew our target out of the water.
Once you’ve set a firm target goal and have decided on a game plan, it’s all about taking action. Start split testing right away and run a solid link building campaign.
Without defined, specific goals, we are subject to the ebbs and flows of life. Once you have a direction to head, then it just comes down to taking action to get there.
Always keep learning
This is just the beginning of CRO as there is a never-ending set of website aspects to test.
Happy with how the buttons are converting? Cool, now test the images on the page to which images build better rapport with the users. Will a video increase customer engagement and keep them on site longer?
If you are really serious about CRO, I highly recommend you check out a book from the guys at Optimizely called A/B Testing: The Most Powerful Way to Turn Clicks into Customers. Once you understand the concepts delivered in that book, you’ll have a solid foundation.
As you can see, split testing is a simple strategy to build off the back of your SEO. Once you have a profitable website, you can add CRO to your arsenal and increase your conversions, but it takes discipline and a clear plan of attack. Ignore CRO and you’re simply leaving money on the table.