While traffic may be on the up following your effective SEO, paid search or content strategies, that doesn’t mean your job as a marketer is done. The people that are visiting your site may not be achieving the goal you want them to complete in the journey. Put simply, they’re just not converting. Converting is a generic term which could really mean anything! It could be a purchase, joining an email list, the creation of an account or even an app download. Regardless of the ultimate goal, a conversion is when a specific action is successfully completed. Before moving on we need to cover an important metric - Conversion Rate (CR). This key metric demonstrates the percentage of your site’s total traffic completing a specific action.Therefore, a high conversion rate is always great (with the assumption being they are the real users). Now, CRO or Conversion Rate Optimisation is the process of making changes to your website, app, form etc. with the aim of providing a more optimised user journey to increase conversion rate. When considering CRO there are a few essential areas that companies need to consider when planning a testing strategy. 
  • A/B and multivariate testing: choosing the correct test type is key to gaining the desired insight.
  • Standardised testing structure: standardising your testing procedure helps ensure statistically accurate results that are easily comparable.
  • User behaviour insight: understanding the qualitative and quantitative behaviours of your users means you can create the tests that are likely to have the greatest impact
  • Text optimisation: on site messaging is one of the key areas that you can influence your visitors - so choose your words wisely.
  • 1st party feedback: Use the resources you have available. Getting feedback from your users is an excellent way to understand them.
  • User segmentation: different types of users will use the site differently and so should be measured separately. Segmentation is vital to identifying these variances in behaviour.
 We cover a few examples in more detail below.

A/B or multivariate testing

The main question to answer here is simply what is an A/B testing? As a top-level example, you set up various landing page experiences where each variant has a different version of an element when compared to the others. For example, you may have a CTA which has the text “Quick Quote” but you hypothesise that based on experience “Free Quote” may perform better. The only way to find the best option? To test it! Using software such as Google Optimize or VWO you are able to seamlessly run tests on your website. You can split the traffic between your variants evenly (50:50) or even just send a test a sample (80:20) if you are uncertain of the outcome. Of course, the text is not the only element you can test. If it’s on the page you can change it! Headlines, product copy, image size, layout, amount of text, fonts. If you have an idea of how an element can be changed to increase conversion rates, it should be tested. The extension of A/B testing is multivariate testing. This means changing multiple elements of one page and dividing your traffic to test every possible combination of your changes. To gain statistical significance however, your site requires a large amount of traffic to test the larger number of combinations with a strong level of accuracy.

You will never achieve the perfect site. Testing should never be complete as there is always room for improvement. Even minor improvements are beneficial to your overall goal.

User Behaviour Insight

User behaviour and how they use your site is becoming increasingly unpredictable. Their journey can be made up of various stages which are important to acknowledge. At Semetrical we consider five stages: 

Strangers > Visitors > Leads > Customers > Promoters

 These user segments will allow you to focus your testing efforts and gradually optimise your site bit by bit. Users can enter your funnel from anywhere, this can be from being advertised to, from researching your product, visiting your website or store, even from the user contacting you. Optimising all of these touch points one by one will enable you to be efficient and structured when testing your site. With all forms of activity, offline and online, it is becoming imperative for businesses to understand their users and how they are moving from stranger to promoter. The collection of data from every channel is crucial when aiming to understand your customer journey. With that being said, it is also important to have the right expertise to create an analytics infrastructure to make this analysis possible and furthermore action it.

Standardised Testing Structure

Once you’ve analysed your data and planned your variables it’s now time to test! When implementing a CRO strategy keeping records of what you have tested, currently testing and going to test is crucial to being efficient and to gain the most benefit. This is important as you do not want to have a scenario where you do not know the aim or the testing procedure when running tests. Below are some key components to consider:
  • Page template: where the test is being conducted e.g. product page.
  • Device: will the test be all devices or a specific device e.g. mobile only.
  • Primary KPI: what metric are you aiming to impact most?
  • Secondary KPIs: what other metrics are likely to be impacted?
  • Variant summaries: if a variant makes a significant improvement, do you know how to fully implement the change? Or how to explain it to a developer?
  • Results: a summary of results following a completed test.
 These are some top level components but this can easily be expanded to include ratings for ease of implementation, start date, impact rating and even stored code snippets for implementing the final result. This centralised log of information will be invaluable when deciding on new tests or reporting to superiors. You do not want to duplicate tests or have missing date.

What else can CRO offer?

Regular experiments leads to more than just an increase in conversion, but It will also lead to a better understanding of your user experience and how you can improve further. Removing unnecessary clicks, simplifying forms, clarity of navigation, all of these can be tested and improve other metrics such as return users or other UX metrics (bounce rate, average session duration and pages per session). Many believe CRO is manipulating the user journey to do what you want, but in reality It’s easing the journey of those users that are already engaged. You are aiming to make their lives as easy as possible so they reach your aim in the most logical but easiest way possible.