Mariam Sharp takes a closer look at some of the tools that can help you with gathering and analysing data

Jérôme Hiquet, Club Med’s VP of Marketing, North America & VP of Marketing & Sales, Mexico, recently highlighted the importance of data to the growth strategies of Club Med. Taking the lean approach of test and learn, Hiquet sees no point in taking on a huge project and then in one year re-analysing what that has meant to the business. Instead he recommends regular AB and multivariate testing that it is possible to take action on with small, purposeful steps.

As a follow on from 5 steps to making a start with A/B testing here are five tools that could help you measure customer interaction. For absolute beginners measuring just one variable is an excellent approach but once you become familiar with A/B testing, you can measure two variables simultaneously (bivariate) or more than two (multivariate).

Stay focused

It’s important to keep in mind that the aim is not to generate data for data’s sake. Stay focused on your business goals and test the items that best inform your approach to those goals. What is key is to make use of your findings to take the actions needed to grow your business.

A good first step is to understand how visitors to your website are moving through the site and where the bottlenecks are. Know the bottlenecks because this can increase the likelihood of potential customers bouncing away from your site. CrazyEgg is a tool that shows you a heat map of where visitors click your webpages. From this information you can draw out the main aspects of your website to adapt or improve.

It is important to keep in mind that the aim is not to generate data for data’s sake

Kissmetrics is another tool that can help by telling you who is on your site and allows you to improve your customer segmentation by creating groups or cohorts. For example you can have a cohort of all the people that signed up in a particular month and data to inform you of their behaviour.

Kissmetrics also claims to help you identify individuals in that cohort, so that you understand how individuals within particular groups behave; and why they might have behaved differently compared to others in the group.

Optimizely offers a simple visual editor for your A/B and multivariate testing. The software makes each element on every page editable and is less complicated than coding individual HTML on webpages. After inserting a single line of code generated by Optimizely into your HTML, you don’t have to touch the code base again. You can also use WYSIWYG editing and real-time reporting.

Unbounce provides a range of templates that help you easily create A/B tests, if you are wanting to focus your tests on your landing page. Unbounce also works with 60 integration partners that include email, CRM, optimization tools as well as a range of others such as Google analytics and Kissmetrics. 

Google Analytics, also referred to as GA, is a free product – well mostly. The tool allows for analytics with numerous add on elements. It also compliments a range of other tools including some of those described above. Google Analytics Content Experiments supports A/B testing for up to 5 individual pages. Also included as part of the free tools in Content Experiments is the multi-armed bandit approach. This approach directs traffic to the ‘winning test’ which can mitigate any losses from pages that do not convert as well.

This is helpful where you run longer-term tests and don’t need to have statistical relevance quickly. The benefit of multivariate tests is that they provide more detail, but are not possible with the free tool. So if you are wanting to do multiple tests, and need to understand statistical relevance, then you will need to try other tools or upgrade to the GA premium version which will set you back $150K per year.

While organisations the size of Travelocity may benefit from working with analytics teams at Google, the majority of benefits can still be achieved by combining the tools above, whatever the size of your company.

Customer concerns

When it comes to gathering data, it is always important to keep customer concerns in mind. The recently revealed controversial study undertaken by Facebook in 2012 to see if altering almost 700,000 users’ news feeds impacted on their mood via ‘emotional contagion’ has triggered both legal and ethical concerns about privacy, transparency and respondent consent.

In the race to understand customers the market for analytic tools are growing rapidly. What are your views on A/B testing and what tools and other methods do you currently use to understand your customer behaviour online? Tell us in the comments box below or join us in Berlin.