Remember the thrill of waiting for test results at school? Whether you were a complete geek and had the excited nerves of a sure-thing A, or a meh-student nervous to see just how much explaining you had to do – there is nothing quite like the anticipation of results.
For thrill-seeking digital marketers and UX designers, the advent of A/B testing is the holy grail of results anticipation. When you combine A/B testing with conversions that equal dollars, your tests become gambles, and there is nothing quite like a win. But if you’re just starting out, how do you know where to start, and what to bet on?
These 12 lessons have been learnt the hard (and not so hard) way from some pretty savvy and creative digital gurus, and will give you the insight you need to roll the A/B dice in a way that will work for your business.
1. Use an above the fold sign up form
For most consumers the first port of call on a businesses website is the home page. For online planning software company – Tom’s Planner – the home page was where they felt their conversions were suffering.
They used an A/B test with the original home page featuring an opt in Sign Up CTA button, vs a homepage with the button PLUS a sign up form above the fold. The results? The variation with the short field form generated 43.85% more conversions.
2. Remove promo code boxes
An ecommerce site featured a promocode box on it’s checkout page. Naturally customers reaching the checkout saw the box and left the page to huntdown promotional codes, never to return to complete the transaction.
Using an A/B test with a variation removing the box, the site managed to increase their revenue by 24.7% throughout the testing period.
3. Use specific CTA buttons
As mass digital consumers most of us have become immune to the temptation of a ‘download’ button. For one online group testing a standard ‘download now’ button against a more specific CTA ‘price guide’, the results were phenomenal.
The variation CTA text with ‘Price Guide’ generated 629.9% more clickthrough’s than the download CTA text. So ditch the download for some more creative CTA testing.
4. Don’t be scared of long form
If you’re one for minimalism and firmly believe less is more, we’ve got some news for you. While each business and market will vary, keeping your webpages (and in particular home page) free from copy clutter isn’t always the best solution.
PayPanther – an online accounting and payment platform – tested their landing pages with a short vs long form A/B Test on their signup page. The longform page included FAQ’s and testimonials and generated a whopping 372.62% increase in signups.
The moral of this lesson is to test test test – giants including Facebook, LinkedIn and Quora still use short form so it’s clearly not a one shoe fits all, but give it a shot!
5. Heading & Subhead copy
If you have a chat to any copywriter about the nuances of tagline and straplines and you’re sure to set off a rant. But with good reason. The copy on your homepage is the first impression your audience gets of your business character – it contains tone and style and can make or break.
For this test you’ll likely need to commit to a longer, multi-variation test to really get a good feel for the words that work. For VenueSphere, an online platform that matches customers with the perfect venue, a change in subheading increased revenue by 69%.
6. Adjust your product image sizing
It seems almost too simple, but sometimes bigger does mean better. While images increasing sales should not be news to anyone, the size of the images used might be.
Online retailer Mall.cz found that increasing the size of their existing product images increased their sales by 9%. The number itself might not seem impressive, but 9% of a high traffic site can be a huge sum of money.
7. Surrender to the ‘free’ gimmick
We use gimmick literally. Marketers know it’s know a ploy, consumers know it’s a ploy, but none of us can resist the little lightbulb that goes on when we see the word ‘free’ next to a CTA or offer.
Online crew Soocial ran a test where they added an A/B variation with a sign up button followed by ‘It’s free!’. The results? A 20% increase in conversions. Not bad. Just be smart and don’t get too scammy with your testing in other areas of digital – for example adding ‘Free’ to an email subject line might do you more harm than good!
Your site is literally your domain to test, and while these are some great lessons, they aren’t going to apply to everyone. Do your homework and map out some areas you want to test – and remember not to test multiple design elements at the same time – control is your number 1 friend when it comes to the A/B thrill game!