Here’s how you can optimise conversions on you website

How many offers does your average landing page contain?

If the answer is more than one, then you’re just like 48% other marketers who cram multiple offers on a single landing page believing that if their prospects have more to choose from, they’re more likely to pick something they like and convert.


This abundance will only confuse and distract them.

Similarly, if your website and user interface are poorly designed, your company will come across as untrustworthy. These aren’t just empty words – 94% of negative website feedback is design related.

But you won’t know any of that, and you won’t be able to improve your conversion rates if you don’t have a well-thought-out CRO plan in place.

cro plan illustration

Know Your Data

The first and crucial step for creating an effective CRO plan is to gather all the relevant data concerning your business, your website, and your customers.

This will provide you with the information necessary for running split tests, which are your No. 1 optimization tool.

The point of this step is to help you understand what you’re measuring and what you want to optimize. There are three areas from which you should collect your data:

  • The company. “Start with why,” says Simon Sinek, and that’s exactly what you should do. Identify what the core belief of your business is and go well beyond your USP and DNA (though these are essential too). Get to the bottom of what makes your business unique and why anyone should buy your instead of your competitors’ products. This part requires a lot of soul-searching, as well as thinking about where you want your business to go and what your goals are.
  • The website. Your website is your main medium of communication with your audience and a vehicle for spreading the message about your brand. This is the place where your prospects go when they want to learn more about your brand so make sure that you impress them within the first couple of seconds in order to make them stay. Check your sales process and review every touchpoint between the customer and the website. Eliminate possible technical issues such as slow load times or mobile optimization. Identify the most common drop-off points and analyze them. For example, if you establish that the majority of your traffic comes from the desktop, it might be a sign that your website doesn’t display properly on mobile devices. And that’s pretty bad given that 57% of people say they wouldn’t recommend a business with a poorly designed mobile website.
  • The customers. This is the part where you should find out what the most common objections of your customers are and what prevents them from clicking that “Buy now” button. The best way to find this out is by asking them directly. There are different survey tools that can help you obtain your customer’s feedback. Not only will you be able to feel your audience’s pulse and eliminate potential obstacles from the customer journey, but you’ll also show them that their opinion matters. The key is to formulate your questions precisely and to offer a limited number of answer choices so that you get the information you need. It’s also important to determine where people who convert come from, that is, how they found out about your company, and identify what channel is the most successful one.

Create a Hypothesis

Once you collect and process your data, it’s time to pinpoint the issue and suggest a solution.

This is the part where you decide on what exactly you want to test. Many attempts to improve a conversion rate fail because people skip this step and proceed directly to testing.

A hypothesis is an educated guess that needs proof, and you have to take all the information you collected into consideration if you want this whole procedure to make sense.

There are a couple of important details to bear in mind.

  1. What will you test?

One of the best ways to start is by analyzing your audience’s most common complaints and objections. For example, the feedback that your “checkout process is too complicated” or that your customers “don’t feel safe when it comes to providing their sensitive data” sends you in the right direction. It’s a no-brainer that you should optimize your checkout process and add some trust signals to your checkout page.

Similar valuable information might be that your customers like your free shipping option, so it’s a good idea to highlight and emphasize it on your product pages, landing pages, social media channels, and ads. In other words, make this selling point more obvious and prominent.

A hypothesis is usually structured in the following manner “If I add trust signals to my checkout page, then my conversions will go up 25%.” It’s important that you should pick something testable as a hypothesis, and in this case, you can measure your conversions and see whether the hypothesis is right.

  1. Who will you test?

It’s only logical that you’ll segment your audience not only based on demographic categories but also where they are in the buying cycle – new and recurring customers shouldn’t be in the same test. This way you’ll prevent unreliable and inaccurate results.

Also, make sure to segment your audience based on the browser they’re using, as that way you’ll eliminate inaccurate results caused by technical issues on a particular browser.

  1. Where will you test?

You can choose to test a single or multiple pages at once. However, it’s worth noting that you shouldn’t test the pages selling 10-dollar shirts and 300-dollar sneakers at once because their buying cycles are different which means that your results won’t be accurate.

Implement Your Changes

It’s showtime!

You can use Optimizely or Visual Website Optimizer to run your A/B tests and compare the results.

Here are a few things that you should bear in mind during this process:

  • Run your tests across different browsers to ensure that the design displays properly.
  • If you’re testing pages that receive a high traffic volume, send only a small portion of the traffic to the test page in case that your hypothesis isn’t correct. That way you even if your conversions drop, you won’t suffer a huge loss in profit.
  • It’s worth mentioning that your conversion rate might not increase, but you should also look at your average order value. If it goes up, then that might be a sign that your hypothesis was correct.

Analyze Your Results

Before you triumphantly conclude that your hypothesis was correct, don’t forget to establish whether your A/B test reached statistical significance.

To put it simply, it’s important to figure out whether the difference in conversion rates between a variation and baseline didn’t occur randomly or by chance.

But don’t worry if all this sounds too technical because most split tools tell you whether statistical significance has been reached.

If your hypothesis was correct, you can deploy the successful design and send your entire traffic there. However, keep an eye on your conversions and see whether there are any fluctuations.

In case your hypothesis failed (yes, that happens more than you can imagine), go through your data again and try to find what went wrong. You can learn a lot from the whole process and use the data for future reference.

The trick is not to give up.

This simple 4-step process can help you create an effective conversion optimization plan and give your conversion rates a boost.

Share This Post

Leave a Reply