The goal of conversion rate optimization is to improve the performance of your website. It is in many ways a scientific method. When you observe how visitors interact with your site, you can create a hypothesis about the areas you could develop and conduct tests to see if those ideas could work.
In this post, we will walk you through this process, give you some tips on how to get started, and hopefully in the end will be able to convince you about the merits of trying it.
Why should you start CRO?
It is easier to convince your existing customers to spend more money than it is to find and persuade new ones. Wingify, the developer of Visual Website Optimizer, carried out a survey among its users. The results show that successful website tests can increase the average revenue per unique visitors up to 50 percent.
Of course, the exact amount of increase you can gain from CRO depends on a lot of factors and changes from industry to industry. But it is very likely that by investing resources in conversion rate optimization, you will be able to make more profit from the same amount of traffic. And since optimizing your site will also teach you what kind of people spend the most on your services, you will also be able to target your advertisements more precisely. In other words, it lowers the cost per acquisition (CPA) too.
The price of conversion rate optimization is minimal, especially compared to the increase in revenue it can bring. On average, companies spend $92 per visitor on driving traffic to their site, and only $1 on optimizing the visitor’s experience. This not only shows that conversion rate optimization is among the least expensive methods of online marketing, but also that it is still somewhat underrated.
In some cases, it is possible to increase sales on a website without researching the visitors’ habits and experimenting. But with conversion rate optimization, you can ensure that your success is not based on outside factors and you can gather data that will let you improve your site even more.
There’s one more reason why this method is beneficial for digital marketers. Websites can be viewed as display windows for organizations, and as a result, at many firms, content is determined by the executives’ taste instead of the marketing experts’ opinion. By applying a scientific method on website development, firms can save themselves from many unnecessary arguments and focus on the results.
When will you need a conversion rate optimization strategy?
Conversion rate optimization is about developing your site at its most important areas and finding ways to convince more people to sign up or buy products from you. While you are looking at factors that could discourage people from shopping, you might also discover flaws in your product. However, conversion rate optimization is not product research. It can only help you increase the chances of selling your products online, if they are, well, sellable.
Other than that, you should start optimizing your page as soon as possible and never stop working on it. Websites always have room for improvement, and no matter how well optimized your site is, it will underperform in some areas.
In the beginning, it’s possible that you will not have the resources or the traffic needed to carry out some of the website tests in a meaningful way. Even then you can run usability tests that will give you valuable insights.
A simple way of doing it is to find people from your target audience and ask them to perform tasks on your site that you would like your visitors to do and write down their experiences. You can also find several tools online that let you carry out remote user tests and help you decide whether your navigation makes sense or if your posts are easily readable.
One of the reasons conversion rate optimization is so valuable is because different things work for different products, and it can show you how your audience will react to various elements that performed well on other websites.
With that said, in the early stages, it is not a bad idea to follow industry best practices or ask for expert opinions so you can avoid wasting time on testing things that are obviously not going to work. Of course, the first person trying out new elements on your site should be you, and there’s a chance that you will find areas to improve by simply going through your checkout process again or reading your content twice.
After a while, most sites reach a stable amount of traffic and conversion numbers. At this period, conversion rate optimization will become one of the most effective tools to increase your revenue.
What to test?
Every element your visitors interact with can have an effect on your site’s performance. When you are looking for areas to test, you should consider how large an impact a chosen area can have and how difficult it is to change it.
Google Analytics can inform you about your visitors’ journey toward conversions and help you define the pages that you need to improve. To generate reports about your sales funnel, you will need to set up goals in Google’s application.
You can do this by clicking on its Admin panel and selecting goals from the options. You can define them several ways in Google Analytics. For the reports to work, you first need to select manual setup, then add the URL that visitors reach when they complete a conversion, and finally, turn on funnels. This will allow you to define the steps toward the goals.
So for example, if you have a subscription-based service, add the URL of the thank you page as your final goal, and add the URLs of the subpages where visitors can learn more about your service as steps. If your checkout process is divided into pages, make sure to add them as separate steps, too. If you have a one-page checkout, it’s a bit more difficult to measure the performance of various parts, but there’s a workaround for that, too.
The reports generated this way will show you where your visitors leave your funnel. For example, if you see that fewer visitors move from the introduction of your services to your sign-up page than expected, you will know that you should focus on creating more engaging descriptions. When you see a huge bounce rate on one of your pages in the checkout process, that is also a very clear indication of where you should take action.
Google Analytics can also help you identify your most popular pages and the ones that produce the highest bounce rate. Both offer great opportunities for optimization. For example, you can come up with recommendations that encourage people to stay on your site a bit longer or create ways that direct visitors to product pages from the well-performing posts.
Using Analytics is the most widespread method of collecting information about visitors’ behavior, but far from the only one. Clicktale and Crazy Egg can generate heatmaps that let you see which are the most clicked or touched items on your subpages. Springtab can help you understand what type of people interact with your site and find out who your most active users are. If you would like to know more about CRO tools and integrating them into your site, read our earlier post on this topic.
Besides analytical tools, you can also ask your customers directly if they experience any difficulties while browsing your site. With Qualaroo, for example, you can create online surveys that are tied to the visitors’ actions and pop-up when visitors try to abandon the shopping cart or search for an item on your site.
Live chat with your visitors or asking customers why they decided to purchase on your site can also provide you with valuable information that you wouldn’t be able to obtain from statistical data. Listening to your consumers will also help you create more persuading descriptions that answer their problems directly.
After you identified the critical areas on your site, it’s time to create different versions for them and test if the changes can lift conversions. Expert opinions differ on whether you should test bigger changes or minor transformations first. Small changes, like applying a different color for your call to action button, are less likely to increase conversions with a very high percentage, but added together small improvements can be just as effective as a complete redesign.
Quick Sprout advises you to test bigger changes first and improve on what performed best, which can work well if you have the resources to carry out the bigger changes. If you are not prepared for that right now, starting with smaller changes can work as well since you will very likely have to change the overall look and feel of your site in two years anyway, and successful changes can carry over.
If you’re looking for areas for testing, the survey from Wingify we cited in the beginning of this post can give you some ideas on where to start. The company also asked its clients which elements were the ones they experimented with first. Thirty percent of their clients tested the call to action button before anything else, which was followed by the headlines (20 percent), layout (10 percent), website copy (8 percent), and the checkout pages (5 percent) as the first choice of element to be tested.
It’s worth noting, though, that besides their usefulness, the ease of changing these elements could also have made these areas popular. Tommy Walker, a marketing expert at Shopify, believes that you can gain the most from conversion rate optimization if you’re not only using the test to increase the percentage of purchases on your site, but also try to understand why your tests failed or succeeded. In other words, if you can discover the problems of your visitors, it will help you more in the long run than the increase in conversions you can expect from the test period.
How to carry out the test?
Besides usability tests and consumer surveys we mentioned earlier, there are two variants of website test that are widely used, A/B and multivariate tests. The latter allows you to change several elements of a page and measure how they interact with each other. For example, you can create three variations for the headline, call to action, and the main image of a landing page and find out which combination performs best.
Contrary to what the name suggests, you can also create more than two variations of a page for A/B tests, but this type of test won’t show you how certain elements interact with each other within one page. These tests only let you know which one of your designs is the most effective.
Since multivariate tests are more complicated to set up, need a huge amount of traffic to bring relevant results, and not something you would want to start with if you are new to conversion rate optimization, we are going to concentrate on A/B tests now.
There are several tools available online that let you run A/B tests. Here are a few we recommend.
If you are looking for a free solution, you should try Google Analytics. You can set up tests within the application by clicking on Behavior and selecting Experiments. Here you can configure what you would like to measure; for example, you can test the goals we wrote about earlier or the bounce rates of your forms. You can also define how long you would like your tests to run, or what percentage of your users should take part in the experiment.
After this, you will need to add the URLs of the original and the alternate pages, and insert the experiment code below the head tag of the original page. You can test up to 10 different versions of a page with this method, and Analytics will notify you when a winner is found.
The downside of using Analytics is that you have to code or ask your developers to create the different variations of the subpages for the tests. Most subscription-based tools do this job for you, therefore using them can be more cost-effective if you’re not into coding.
Unbounce lets you create and test different variants of responsive landing pages. You can use its drag and drop tool to edit your test pages or choose from several templates. It is easy to integrate the application into WordPress websites, and you can connect its forms to your HubSpot, Salesforce or MailChimp accounts, too.
Visual Website Optimizer also lets you change and test elements on a page without coding. It provides heatmaps, too, that can help you decide which areas to test, and allows you to personalize some aspects of the test pages. For example, you can show different versions of your websites to visitors based on where they coming from or what type of device they use. The application can work together with most of the popular content management systems and Google Analytics.
AB Tasty also lets you create the subpages for your test with a visual editor. Besides conversion rates, you can measure several other goals with this app from engagement to average cart value. It also helps you to segment your audience and create personalized campaigns.
There are several other tools that you can use for A/B testing, you can find a longer list of them in this post.
No matter which solution you end up using, if you follow a few tips we collected, your test results will most likely be useful and relevant.
- To make sure your testing software is working properly, run an A/A test before you start the real tests. While there could be some minor differences in the performance of the pages, if everything works as it should, they will even out after a while.
- Run your test until you get statistically significant results. The industry standard for A/B tests is 95 percent, which means you should consider the test a success if the alternative page has a 95 percent chance to perform better than the original.
- Test your pages for at least a week. People use the internet differently on the different days of the week and this can affect the results. For example, a B2B site usually has fewer visitors on the weekend, and a testing in this period can be misleading.
- You don’t need to worry about tests influencing your ranking if you use one of the most popular platforms. However if you create your alternative pages, read Google’s guidelines for A/B testing, and follow the described steps to make sure the test won’t have a negative effect on your SEO.
- If the aim of your A/B test is not in preparation for a major redesign, but you are more interested in the effects of smaller changes, don’t test more than one element at one time.
- When you analyze your tests of several elements, concentrate on the end result. For example, if you change a picture of a product and it grows the number of clicks but doesn’t lead to more sales, you shouldn’t consider it a success.
See below for some of the sources we used while creating this post, and the articles we recommend if you’d like to read more about conversion rate optimization: