Data-Driven Marketing & Conversion Rate Optimisation
Companies who see improved conversion rates are performing, on average, 50% more tests and use 47% more methods to improve conversion.
Conversion rate optimisation is the process of updating and testing elements on a webpage with the ultimate goal increasing the percentage of website visitors who take a desired action. As the process is based on conducting experiments, being able to collect and analyse data is key.
Setting Up for Your Experiment
There’s no point starting a Conversion Rate Optimisation campaign without having the tools in place to collect your data. Google Analytics is an excellent starting point. In addition, you may want to look at the information in your customer database, CRM and any automation platforms that you use.
These platforms should be set up in a way that segments the information correctly, as this is key to informing your online strategies. For effective testing, some of the most important pieces of information to consider are:
- The pages or sections of your website that you want to test.
- The audiences that you want to be included or excluded from your experiment.
- Any valuable actions that indicate a change in status for a user on the site, such as clicking a downloadable resource or visiting a certain page. This is most relevant for lead scoring.
- The specific channels that drive traffic to your website.
Depending on your business and objectives, there may be other things that you want to segment and track. It’s important to consider these at the start of the campaign so you can properly set up tracking from the beginning, rather than realising halfway through and having incomplete data.
Identifying Test Areas Based on Data
Before testing, it is vital to look at your data, as this will set a benchmark for your experiment. Use your segmentation to look for patterns of performance, combining data from the different segments to identify actionable areas of improvement.
It’s important to remember there are a number of things that can affect the performance of your digital campaigns. If your website has a variety of landing pages and one page has a much lower conversion rate than the others, you may jump straight to the on-page elements such as design, forms and calls to action. One of these elements could well be the problem, however looking at the bigger picture of the campaign can provide more in-depth insight. Consider looking at the data from different segments such as the traffic going to that landing page from different channels, or devices.
By looking further into the data in this way, you can get more information that may inform your wider campaign. For example, if 80% of traffic to that particular landing page was coming from a paid search campaign, you may have found the problem. In this case, you’d have to review both the ads and the landing page to understand how they align and pinpoint the cause of the low conversion rate. This could involve:
- Testing new ads, and looking at click-through rates as well as conversion rates from traffic to other landing pages
- Testing a new landing page, and looking at all traffic channels’ conversion rates
Don’t test both of these things at once. Choose one to start with, and start conducting the experiment to see if it makes a difference.
Running and Analysing your Experiment
Once you have determined what metrics you want to analyse, and you’ve set up your experiment, most of the hard work is done. Your experiment will run and you will start collecting the data that you need to optimise your conversion rates and boost your overall marketing campaigns.
The key thing to remember once you have your data is to ensure you analyse it objectively and don’t fall into the trap of confirmation bias. This is when an analyst subconsciously looks for information that confirms their hypothesis and places more weight on this data than on data that disproves it. This can lead to statistical errors that can be costly for the business.
It’s also important not to rely too heavily on software programs to analyse the data for you. While programs can provide the data, they can’t fully explain and interpret it in order to help make marketing decisions that drive business goals. Data analysis is about creativity as well as logic and requires a combination of the right tools as well as a human touch.
“Marketing and data science are only just getting acquainted. There will come a time when analytical techniques are built into most workflows and machine-driven decisions are commonplace. At such a time, data science will no longer be a separate activity but the essence of marketing.” – Gartner
Today’s marketers should be familiar with the relationship between marketing and data science to keep their marketing strategy on top of its game and ensure they are optimising conversion rates in the most effective way possible.