Conversion World 2015 is over. We’re all exhausted and happy, not least the the team behind it, Digital Tonic. This was the first event of its kind, an ambitious, international, entirely online conference on conversion rate optimisation.
If you didn’t attend here’s a rundown of the best bits you missed.
Advice for ecommerce sites
Break down the funnel stages
Separating the data for different parts of your site will help you identify the best and worst performing parts of your site.
In the example below cart abandonment rate is very high. 75% of shoppers started the checkout but almost 89% of them left the site without buying anything. The best place to start optimising is the most broken place, the checkout.
Work on your implicit codes
Implicit codes are signals that convey straight away what kind of site it is. In the case of an ecommerce store it should be clear that you sell things, what things you sell and who you sell them to.
Make your site search-centric
John Ekman, @, conversionista.se, investigated the theory that consumers who use site search convert better. In his case study, users who searched had a 91% higher conversion rate, a strong indication that internal site search is good for conversions.
We can assume that users who search have a greater intent to buy than those who don’t. However, a great internal search feature has the power to encourage conversions. Allowing users to filter your products means they get to what they’re looking for faster.
Do your research
Hazjier Pourkhalkhali told how at Optimizely they seek to understand as much about their users as possible. This means understanding customers needs, aspirations, expectations, fears etc.
Next, gather data on users interaction with your site:
- Acquisition channels (organic search, paid campaigns, social media) versus conversions.
- Devices: traffic versus conversions.
- Time to purchase i.e. how long between first visit to purchase.
How to get personalisation right
Xavier Colomes presented a session on personalisation. Although personalisation has been a predicted trend for at least the last 10 years it’s yet to really take off, something marketers seem slow to realise. Only 37% of consumers say their favourite retailer understands them.
Colomes explained that he sees very few examples of true personalisation. Segmentation by new and existing user doesn’t go far enough. We need to learn more about our users.
Colomes says to encourage more users to register and log in to your site you have to show them the value of doing so.
Here’s his three steps for achieving successful personalisation and how to get buy-in from your users:
- Personalisation – experience as value. Improve your users’ experience using the information you gather. Filter the content to show the most relevant products or content first and personalise your messaging to create familiarity between you and the user.
- Collaboration – recognition as value. Recognise what actions your user has taken already e.g. know which whitepaper they’ve already downloaded. Think also about how Trip Advisor give you a pat on the back for helping other travellers with your reviews.
- Manipulation – saving as value. Offer discounts for everybody, rewarding loyal users as well as attracting new ones.
Look out for a more detailed post on personalisation and 1:1 marketing on the blog soon.
How to hone your unique value proposition
Continuing on the subject of persuasion, Rich Page’s presentation “How to optimize a hidden conversion killer: Your Unique Value Proposition”, contained lots of tips on perfecting your messaging.
A weak unique value proposition (UVP) can kill off a conversion in the first 10 seconds of a visitor landing on your site.
Get it right by explaining, in the most appealing way possible, these 3 things:
- What you offer and to whom.
- The benefits of what you offer.
- Why the customer should buy from you over a competitor.
This might seem like marketing 101 but the point Rich makes is that from within a business there’s often a can’t-see-the-wood-for-trees situation. Don’t ever assume visitors to your site know what you do or why you’re great already.
UVP must always be customer and benefit focussed, like ao.com. It’s pretty clear straight away that AO do deals on appliances. The chance to make savings is their biggest selling point and they reiterate it frequently.
Reiterate your UVP in the sign up and checkout pages where abandonment is usually highest. Put emphasis on ‘risk reducers’ such as secure checkout, free returns, money back guarantee to answer the shopper’s worries at this critical point.
Page and Wiebe recommend that you look at what your competitors are doing but be wary of copying. Your site is different and what works for them might not work for you so always be critical, always test.
Testing, in particular A/B testing, was a major theme of the conference.
It’s so important to test and gather data because making changes based on what you think will work better can be a waste of time. Make decisions based on what your own site is telling you, rather than simply following what your competitors are doing or what you’ve heard is ‘best practice’.
Avoid blindly adopting best practices
Karl Gilis (@) addressed the danger of implementing best practices for the sake of it in his presentation “Best Practices Don’t Work”.
Some of the myths busted:
- “Sliders (aka carousels) suck”. Actually they can be okay when they’re not used to give multiple messages.
- “Put reassuring messaging near to a call-to-action” to generate more clicks. But you must consider your users and the wording of your message.
- “Green check marks always work!” Gilis showed several examples where adding green check marks didn’t affect conversions at all.
- “Tell users what (you want them) to do”. Gilis showed some surprisingly big increases in conversion doing this and some examples where nothing happened at all.
Use tools to find the problem areas and improve those, testing your ideas before rolling out changes. Gilis uses Google Analytics, Morae, MyGaze, Tobii, Mouseflow, Inspectlet, Hotjar, CrazyEgg, Formisimo and Qualaroo.
What to test and when
Paul Rouke discussed the breadth of optimisation strategies. For anyone struggling to decide where to start optimising or if you’ve come to stand-still in your optimisation process, this is for you.
Simple changes include headline changes, layout changes and form field optimisation. All changes should be based on test data, rather than guessing at what will improve your site.
Small changes usually achieve small uplifts. Rouke’s case study of a small testing project achieved 4.7% increase in conversions. There’s less risk and less effort involved in simple testing. In addition to the uplift in conversions simple testing is a great opportunity to learn testing skills. It’s also great for learning about your users and what works for them.
Radical changes include complete redesigns, changing your business proposition and reducing steps within a flow e.g. checkout steps.
Rouke clarifies that the aim of radical testing is to grow your business not just optimise your website. The radical testing case study shown achieved 78% increase in conversions.
Visit Rouke’s presentation slides From Simple to Radical: Exploiting the Full Spectrum of Testing Opportunities.
And for more insight on letting simple testing inform radical changes read Evolutionary vs Revolutionary Site Redesign; Why Not Both.
Excusing some technical hiccups, the concept and the conference itself were inspired. We’re well into conference season right now on both sides of the Atlantic, with many keen optimisers flitting around the world. Conversion World stands out as a location-free event. It facilitated talks by the best speakers on a platform that wasn’t limited by location, cost or travel requirements. Some attendees dubbed it ‘The conference to attend in your pyjamas’, basically easy-peasy.
The greatest appeal was the quality of the speakers and the value of their session topics. Each of the speakers offered high value tips with plenty of real world examples.
For more tips on conversion rate optimisation sign up for email alerts below.