A/B testing tends to confuse most of us for not other reason than it sounds like something you need a science degree in. Fact is, A/B testing is the simplest thing in the world. It’s the age-old test of giving people two options and seeing which one they like better. It’s also a much cheaper method of knowing what your clients want without investing in a pricey focus group.
Often it’s little tweaks that can be made and tested through A/B. Do your clients prefer a website or email template layout A versus B. It could be something as simple as preferring navigation options on the right side of the page versus the left, but the design preference could mean a higher traffic numbers and click through rates. Typically, there’s a check list of elements that can and should be tweaked, regardless of whether you’re testing an email or a landing page. These include:
- “Call to action” buttons including language and graphics:
“Follow us on Twitter” works better than “We’re on Twitter.
- Form fields:
Short form fields do better than long ones, even when fields are optional on long ones.
- Layout and design:
Right-hand placed content and buttons do better.
- Headlines and taglines:
Concise Journalism style headlines get more click-through rates.
- Images and graphics:
Consistency in style yields greater conversion.
- Language tone and length:
Short and conversational copy is often preferred to longer copy.
The above methods are traditional approaches. To think beyond traditional parameters, reconsider not just how you treat your testing group. For a winning A/B strategy, reconsider how you treat your product.
A conservative mind considers a product as a commodity that can only be bought or sold. A mind that’s thinks outside the box knows that whether you’ve got a tool or software as a service product (SaaS), it can go beyond just a flat product. These types of thinkers understand that users should be able to “test drive” software. It’s an advantage users are increasingly becoming accustomed to considering the high rate of trial periods for online tools and software. The fact is that users more likely to convert once they’ve opted in for free. In some cases, figures showed a +100% conversion rate. 37signals for example, found that offering a 30-day free trial on all accounts yielded a 30% increase in conversion rates versus their original language of “start a Highrise account.”
Of course, you’ll have to consider that some users opt-out during for after the trial period, though not remarkably enough to not warrant this worth your time. Getting an increased number of trial users also gives your company an increased user pool to run your ongoing series of A/B tests.
Free trial periods also bring up the tricky subjects of data sharing. Data sharing usually makes or breaks a potential conversion. Here’s where a couple of simple solutions can increase your yield. Call them “trust symbols”, these icons can make or break a conversion. Consider this: would you rather purchase on a site that has a trusted PayPal logo or one that doesn’t? Of course, we all know PayPal is a secure logo, but are there other trust logos? What about information and data sharing privacy measures? Does a client feel comfortable handing over their contact info knowing that information wouldn’t be passed along to a third party? Most clients don’t consider this, but conversion rates for actions such as requesting a quote or entering in private contact info do increase when there are “trust logos” in place.
Also keep in mind that when offering a free trial, users tend to shy away from a sign-up process that requires coughing up your credit card information up front. In a conversion study by Totango entitled “2012 SaaS Conversions Benchmark”, the company examined several business models that involved 10,000 unique visitors over a 90 day free trial. The results were staggering. Considering a “credit card required” to “no card required” ratio, sign-up conversion rates varied by 2%:10%, with the latter having a 20% retention increase after a trial period expired.