Chapter 21: Lessons Learned from 21 Case Studies in Conversion Rate Optimization
Editor's Note: Conversion rate optimization (CRO) must be tailored to support the specific goals of a website's business. After observing more than 1,000 split tests for landing pages, however, Paras Chopra identified a number of general patterns among the ones with the highest conversion rates. He shares them in this post, which was originally published Aug. 3, 2010, on The Moz Blog.
Conversion rate optimization (CRO) is the newest darling of Internet marketers. After all, what good is traffic if it doesn't convert? Unfortunately (or fortunately, depending how you look at it), unlike pay-per-click (PPC) marketing, CRO isn't a game of how much money you can throw. In fact, this field requires as much creativity as it requires monetary investment. That's what makes conversion rate optimization a fair arena. Your well-funded, bigger competitors can of course beat you at generating more traffic, but they can't beat you at the conversion rate game (unless you allow them to).
Every website has unique conversion goals, so the best approach to CRO is unique for every website. You should not expect to follow tips from a “best-practices” article and boost your conversion rates instantly. Chances are high that what worked for others may not work for you. So, the biggest step in increasing conversion rates is coming up with creative ideas and designs that can work
Even though CRO is a highly customized process for every website, over the course of the last couple of years (and the course of more than 1,000 split tests), I have observed a few general patterns that yielded great results. Different ideas for increasing conversion rates are worth discussing because they become a great source of input for coming up with your own ideas. In this chapter, I discuss a variety of ideas for CRO, detailed through different case studies. Four main elements that influence conversion rates are as follows:
• Design
• Social proof
• Headline and copy
• Call-to-action (CTA)
Let's start by discussing the role design plays in increasing conversion rates.
The Role of Design
From a conversion perspective, the design of a website is the most important aspect among all variables involved. The difference between a more successful converting design and a less successful converting design usually boils down to making it clear to the visitor what he is expected to do on a page. Take a look at the following examples, showing the old and new site design for Basecamp.
What made the new design convert 14 percent more visitors? A clean design. The new design clearly guides a visitor towards the Plans and Pricing button, while the old design presented a whole lot of choices.
Basecamp home page design: 14 percent increase in conversions
Need more proof that having fewer choices on the page can increase conversion rates? Have a look at the following case study results (http://mz.cm/166nB5h).
Gyminee home page redesign: 20 percent increase in conversion rate
You can check out other similar case studies where design played a key role in optimizing conversion rates at the following links:
• How we increased the conversion rate of Voices.com by over 400 percent (www.conversion-rate-experts.com/voices-case-study/)
• Skype home page redesign (http://mz.cm/XR21hy)
• Official Vancouver 2010 Olympic Store home page redesign (www.getelastic.com/ab-test-case-study-homepage)
• Performance Based Design—Web Design Book: 131.2 percent improvement on landing page (www.abtests.com/test/83001/landing-for-performance-based-design-web-design-book)
The Role of Social Proof
In addition to reducing the number of choices for the visitor, having a design that demonstrates you are a professional and trustworthy company can also increase conversions. Take a look at the following case study, where the redesigned sales page has various trust elements (seal, money back guarantee, testimonials) and the design has various little tweaks (color scheme, buttons instead of links for download, layout, etc.) which made it look professional (see http://visualwebsiteoptimizer.com/split-testing-blog/how-aquasoft-increased-their-sales-by-20-doing-ab-split-tests-in-multiple-phases). Note that sales (not just conversions) increased by 20 percent when the design was changed. No additional products, no additional traffic, pure CRO.
AquaSoft sales page redesign: 20 percent increase in sales
Or take another case study, where simply replacing a “Never Beaten on Price” image with an “Authorized Seller Site” increased sales by 107 percent. The conversion rate increased from 1.81 percent to 3.76 percent, effectively doubling Express Watch's business (http://visualwebsiteoptimizer.com/split-testing-blog/increase-in-sales).
Express Watches replaces “Never Beaten on Price” with “Authorized Dealer Site”: 107 percent increase in sales
Social proof has shown to increase conversions on repeated occasions. Wikijobs.co.uk had one goal when they decided to A/B test how customer testimonials affect conversions: to increase sales (http://visualwebsiteoptimizer.com/split-testing-blog/customer-testimonials-increase-sales).
Wikijob.co.uk add testimonials: sales increased by 34 percent
Adding customer testimonials increased sales by 34 percent in spite of the testimonials being very sober, versus the overly enthusiastic ones you usually see in marketing collateral.
Check out some of the following links on how social proof affects conversions:
• Harvard Business School study finds one-star increase in Yelp rating increases revenue 5-7 percent (www.hbs.edu/research/pdf/12-016.pdf)
• Trust Badge on website results in 32 percent increase in conversions (http://visualwebsiteoptimizer.com/split-testing-blog/do-trust-badges-on-websites-work-oh-yes-32-increase-in-conversions)
• Check out page A/B testing: three dead simple changes increase sales by 15 percent (http://visualwebsiteoptimizer.com/split-testing-blog/checkout-page-ab-testing-increase-sales)
The Role of Headline and Copy
When you receive an email, it's the name of the sender and the subject line that influence your decision to open it right way, or to postpone it. Similarly, when a visitor arrives on your website, it's the design/brand name AND the headline of the page that influence his decision to engage with your page or not. Visitors' attention is the most valuable commodity on the Internet, and your page's headline is where it goes right after arriving on your page.
Take a look at the case study at http://37signals.com/svn/posts/1525-writing-decisions-headline-tests-on-the-highrise-signup-page, where 37Signals tested different kinds of headlines (and the winning one boosted the conversion rate by 30 percent).
Highrise headline test: 30 percent increase in conversions
The winning variation said “30-day Free Trial on All Accounts,” and the worst performing variation said “Start a Highrise Account.” Note that the clear, no-nonsense headline won. If you think about it, if a visitor is on signup page, he obviously knows that he is signing up for a Highrise account. The winning headline clearly convinces the already interested visitor that there is nothing to lose, as they offer a 30-day free trial.
Here's another example of how much headlines matter: CityCliq, a startup in the local marketing industry, split tested the positioning of their product (http://mz.cm/XR4nNt).
CityCliq headline test: 90 percent increase in conversions
Here are the headlines CityCliq tested:
• Businesses grow faster online! (too fuzzy, and so what if they do?)
• Get found faster! (found where?)
• Online advertising that works! (too generic)
• Create a webpage for your business. (clear, concise, and to-the-point)
The winning headline “Create a webpage for your business” tells the visitor exactly what CityCliq does. No wonder it increased conversions by 90 percent. As they say, don't make your visitors think.
Right after looking at a headline, if his interest is piqued, a visitor looks at the (text/video) copy on the page. That's why a combined optimization of headline and copy proves to be effective, as it did for Moz (www.conversion-rate-experts.com/seomoz-case-study).
Conversion Rate Experts' “How we made $1 million for Moz” strategy
They tested a variety of headlines and copy elements on the landing page for a Pro subscription. In the end, they found out that a headline that piqued interest, and copy that laid out what exactly constitutes a Pro subscription won.
Check out some other case studies where headline and copy mattered:
• Dustin Curtis' “You should follow me on Twitter here” experiment—12.8 percent increase in conversions (http://dustincurtis.com/you_should_follow_me_on_twitter.html)
• Marketing Experiments Response Capture case study—a triple-digit increase in conversions (www.marketingexperiments.com/blog/research-topics/response-capture-case-study.html)
• Landing Page case studies from SiteTuners (www.sitetuners.com/landing-page-case-study.html)
The Role of Call-to-Action
So, you optimized your design, your headlines, and your page copy. You got the visitor interested and motivated to try whatever you are offering. There is still one last hurdle before you can throw a success party for your CRO project. Yes, the call-to-action (CTA) is the last hurdle for you to cross. Even though the CTA may be considered as minutiae for CRO (www.moz.com/blog/dont-fall-into-the-trap-of-ab-testing-minutiae), the following case studies demonstrate that even simple A/B testing of CTAs can result in great improvements.
A highly motivated visitor will sniff out even the poorest of all CTA buttons. So while optimizing this aspect of your page, make note that you are optimizing for the busy, semi-interested visitor. If he can't immediately see how to try out whatever you are offering, he will hit the back button. (And in CRO, the back button is the greatest enemy of all.)
37Signals' call-to-action: signups increased by 200 percent
The now-omnipresent “See Plans and Pricing” button increased signups for Highrise by 200 percent (http://thinkvitamin.com/business/how-to-increase-sign-ups-by-200-percent). I have included this case study not to convince you to replace all your buttons with this text (it may not actually work for you). Rather, the point is to convince you that even small changes in the CTA can have a dramatic impact on conversion rates. And the best thing about CTAs is that they are so easy to test. It literally takes five minutes to get such a test up and running.
Another oft-repeated test is to see which color works best for a CTA. (Unsurprisingly, a bright color such as red mostly works better; this may be because bright colors are eye-catching and drive visitor's attention towards them.) As an example, along with testing “Sign up for free” versus “Get Started Now” buttons, Dmix also tested green versus red buttons—and found out that red works better (http://dmix.ca/2010/05/how-we-increased-our-conversion-rate-by-72).
Dmix case study: 72 percent increase in conversions
To repeat my earlier point, with CTAs, sometimes surprisingly trivial changes can produce significant results. Take a look at the following case study on Soocial's home page (http://visualwebsiteoptimizer.com/split-testing-blog/ab-test-case-study-how-two-magical-words-increased-conversion-rate-by-28).
Soocial's home page: 28 percent increase in conversions
Notice that all they did was to add “It's free” alongside “Sign up now” to boost the conversion rate. This is definitely a trivial change. And why wouldn't you test such trivial changes if it doesn't take much effort and has the potential to fatten your bottom line?
Here are some other case studies where tweaking the CTA helped increase the conversion rate:
• Get Feedback v/s Respond Now—13 percent increase in conversion rate (http://rypple.com/blog/give-feedback-vs-respond-now)
• RIPT Apparel's Buy Now button test—6.3 percent increase in sales (http://visualwebsiteoptimizer.com/split-testing-blog/how-ript-apparel-increased-their-sales-by-6-3-ab-testing-case-study)
• Firefox Download button test—3 percent increase in downloads (http://blog.mozilla.com/metrics/2008/11/21/changing-the-firefox-download-button)
Your Role
The framework of optimizing the design, headline, copy, and CTA of each conversion page should provide you with a good starting point for creating your own CRO strategy. What matters in increasing conversions are two factors: not making your visitor think too hard about what you are offering, and showing them how to actually try that offering. Try to make everything obvious and simple, guiding your visitor from headline to copy to CTA like a smooth-flowing river.
However, no matter how many case studies you read, in the end, your conversion rate optimization program will turn out to be unique because your website is unique, your audience is unique, and your goals are unique. The real key to increasing conversion rates is to keep experimenting and testing what works for you.