Conversion Rate Optimisation
This service is about improving the conversion rate of your website.
It achieves this by “tuning” the content and suggested “conversion pathways” you offer to better match the desires of your website visitor. Think of it as improving your website’s persuasive ability.
We apply this service through a four-step process.
1. Understanding all that we know now
Before the project starts we spend time in your Google Analytics environment. First we need to know whether the conversion rate is being reliably calculated across all the conversion options you provide. Then we dig into your reports to uncover all we can about your visitors; eg, their on-site behaviours and overall demographics. Then it becomes easier to start building a hypothesis.
2. Designing your first experiment
Now we define what your first experiment should be. Think of this as less “I think your header image should be a tree” to more like “80% of your home page visitors fail to scroll further than 25% down your page. So in this test we’re going to reduce the depth of the header image to ensure the testimonial content below becomes visible. By doing this we believe the page value of this page will improve by 20%”.
3. Implementing the test
Now we test our hypothesis against live data. We use a number of tools to apply tests across various website environments. Google Analytics has A/B test capability, and we can also use separate landing page creation tools. The goal is to get the test up and running as quickly and cost-effectively as possible to allow enough time and budget to apply what we learn.
If your website has enough traffic you can move from A/B into Multivariate testing. This is where you run multiple content changes through the test at the same time. The Confidence Level of each recipe is then predicted based on the conversions it has achieved. The more traffic you receive the faster your tests will run.
4. Incorporate the learnings
Once the experiment has completed, we incorporate all the good stuff then plan the next test. Clients who use this service generally have experiments of varying levels of complexity running at any one time. By doing this they are actively tuning their site for ongoing improvement.