Given the fairly unpredictable nature of societal trends and human response, conversion optimization is a remarkably scientific process.
While conversion optimization rules are applied differently from website to website and market to market, the rules themselves remain constant. At it’s core, conversion optimization comes down to a rather simple process: take a set of established principles and A/B test them across all features of your site until you determine which options stick best.
Like I said, it’s a simple process. Apply universal principles and test, test, test. So what’s the problem?
It Takes Time
The reality of testing is that it can be laborious and time consuming. When changing a single word can affect your conversion rate by 161%, nothing is too small to be tested. You could spend the rest of your life testing every possible nook and cranny of your website and still leave money on the table.
So what do you do when you have a set of universal principles, backed by 100+ case studies, and the realization you’ll never be able to fully test each principle?
If you answered, “Test as much as possible and continue testing throughout the company’s existence,” congrats! You can go home now.
Unfortunately, a large number of business owners are simply applying the principles of conversion optimization carte blanche to their sites and then calling it a day. They are increasing the CTA sizes, adding relevant images, offering discounts, posting videos, and lengthening landing pages.
But what if removing the video from your home page could increase your conversions by 35%? What if including the price above the fold could increase sales by 100%?
The fact of the matter is: on your website—in your niche, even—the rules of conversion optimization might be completely irrelevant.
In the pursuit of sales, leads, phone calls, subscriptions, signups, and serenades, the principles are merely a base from which to create quality tests. If you don’ test, following the rules of conversion optimization could actually kill your conversions.
Test, test, test, test, test. If you don’t believe me, here are 7 case studies where results completely broke the rules of conversion optimization.
1) Replacing Video With Image Slider Increases Signups by 31%
If you’ve ever read an article on increasing website conversions, you know that image sliders are the worst! Just look here, here, here, here, here, here, and here. You’ll find conclusively that image sliders do indeed suck. The principle behind this idea is users want to experience your site at their own pace.
Of course, you already know what comes next. After enduring disappointing signup rates, Device Magic changed its home page from a video:
To an image slider:
This one change resulted in a 35% increase in click-throughs to the sign-up page and a full 31% increase in actual sign-ups.
Of course, creating a better video might have produced positive results as well, but when you consider that a universally praised feature was successfully replaced with a universally maligned one, the message is clear. There are no universal rules for your unique website.
2) Removing Relevant Image Increases Conversions By 400%
When it comes to landing and checkout pages, gaining customer trust is essential. Trust-inducing (think lock, key, or smiling face) images are often used as a part of this process.
A case study by Hawk Host found that simply changing the image on their homepage from a globe to a padlock increased conversions by over 200%. Another study by OrientalFurniture.com found that including a security seal image increased sales by 7.6%. The general consensus on images is that they have a decidedly positive effect on conversions…
Except for when they don’t. A case study by a coupon blog found that eliminating a “Secure” image from the sidebar resulted in a 400% rise in click-throughs.
Notice I didn’t say 4%. Going against the principles of conversion optimization increased click-throughs by 400%. Still comfortable with blanket application?
3) Adding Price Above The Fold Increases Sales By 100%
An old sales adage emphasizes one should never reveal the price of a product or service too early. As Moneycation puts it, “Price is irrelevant until you have established value.”
Traditionally, “too early” meant anytime before the final part of the close. The price must remain hidden until the last possible moment, or you will scare your customers off. In the digital spectrum, this looked like a price-absent landing page requiring a positive click-through in order to reveal pricing.
While this rule is not as strictly observed by digital marketers today, it is very common to see prices buried at the bottom of a lengthy landing page. A case study with SafeSoft Solutions, however, found that prominently including a price above the fold increased conversions by 100%!
As you can see, there is zero subtlety in this test. The price is bright, bold, and beautiful, and the results speak for themselves.
Is leading with the price the best option for your website. Probably not. But it could be. Test!
4) Adding An Offer Nobody Wants Increases Sales By 233%
This one is a bit out of the box, so I wanted to include it on our list. When it comes to pricing, it seems people are predictably irrational. According to concepts detailed by Dan Ariely, customers can be influenced into making rational purchases with the inclusion of somewhat irrational, inferior offers.
The best way to illustrate is with an example, so we’ll jump straight into a case study. This study was published on Unbounce and started with the testing company’s two primary service offerings, shown below.
Obviously, the company’s preferred choice was to lock clients into a yearly program, but using the above model, they had 60% of orders in the monthly category and only 40% yearly.
Using Ariely’s ideas on customer behavior, the company added an irrationally inferior offer and stripped away the full round of features on all but the intended yearly plan. Take a look at the new model below.
As predicted, adding this completely unreasonable offer increased the percentage of $299 yearly sales to 86% of the mix, resulting in a 233% overall increase in conversions.
The idea is to add an inferior product that highlights the value offered by the product you actually want to sell, and it works like a charm.
5) Increasing Prices Increased Revenue By 114%
We all know that price point is a tricky game, which can swing in either direction. That said, we tend to believe lower prices equal more sales and higher overall pay. My point with the following case study is to encourage business owners not to underestimate the value of their products. It’s very easy to do, and it can cost you big.
Server Density ran a case study on their price point. They started with an extremely broken-down pricing system, allowing customers to pay per server and website, starting as low as $13 per month.
For the test, they decided to implement pricing packages, drastically raising the lowest-priced offer available to users. The new lineup looked like this.
$99 per month is a significantly higher entry price point than $13. You might think conversions would have tanked, but perhaps the per-unit profit would be sufficient for a slight increase. Interestingly, paid conversions dropped by only 25%, resulting in a 114% increase in revenue.
Imagine that. Your customers might value your product at double what they’re paying now. Isn’t it worth a bit of testing to find out?
6) Short Landing/Home Page Increases Signups By 13%
It’s typically understood in digital marketing that longer landing pages convert better. Many experts will tell you that failing to address customer concerns or objections is a primary reason landing pages fail. Crazy Egg increased conversions by 30% with a new landing page 20 times the length of the original! That is not an exaggeration—20 times!
But long pages are a lot of work. What if your site actually needs a shorter page? For Designboost, a shorter page was exactly what was needed. They went from a 6400 pixel long homepage to this:
This much simpler page resulted in 13% more email signups and a 25% higher click-through rate. If your home or landing page is intended as a gateway, shorter may be better.
But the whole point is simply that you can’t know what works for your site until you’ve tested it. As I mentioned at the beginning: testing, testing testing.
7) Removing Social Proof Increases Engagement
For our final showcase, it’s time to address social media. The idea of creating viral content has taken a choke hold on unknowledgeable business owners the world over. Job postings for $15/hour viral writers are littering the blogging boards, as if such individuals actually existed.
In today’s climate, it’s very easy to get enamored with the potential of social media to catapult your business to prominence overnight.
Unfortunately, this idea is as much a fantasy as your high school dreams of playing in the NBA. Many business owners are missing the point of social media, and this misunderstanding is affecting your conversions.
A case study of CalPont’s website demonstrated social proof can actually be harmful for some businesses. The site featured social share buttons above its articles, as seen below:
After eliminating these topside social icons, traffic increased as well as several other measures of engagement. The testing agency found that displaying social shares below a certain threshold actually discouraged users from engaging.
For certain industries, products and services are geared toward shareable content. For others, social media can actually distract and disengage users. Testing!
The Principles Are Simply A Baseline
As I mentioned at the start of this article, conversion optimization principles are fantastic to know if you’re using them as a baseline for quality tests. Blindly applying universal rules could kill your potential revenue by 400%. Ouch!
Test, test, and test some more. It’s as simple as that. Successful businesses never stop testing. It takes time. It’s tedious and meticulous.
But what if you’re one word away from a 161% increase in sales?
Read more conversion-minded articles by Jacob McMillen