Our previous article on the basics of conversion rate optimisation (CRO) looked at:

  • What CRO actually stands for.
  • Why you should use it.
  • How to get started.

This time we’re going to start lifting the lid on how it’s actually done.

If you’d like to get more business from your website, then read on. We’re going to show you some tricks of the trade …

How to do CRO planning and testing

So, at the end of the last article, we’d found our CRO problem. The recruitment agency in our example needed to increase candidate signups for a particular job role.

But how would you apply that? How can you determine what you actually have to do to get more conversions? It’s time to do some more research and write a CRO plan.

CRO research

After defining what you’re trying to improve, the next step is to find out how to do it. How though? By conducting further research on relevant areas of your website of course.

There are a few ways you could do this – with common examples including:

  • Focus groups replicating your customers’ journeys.
  • Heat / click mapping showing actual customer journeys.
  • Data analytics showing the type of traffic you are receiving and why.
  • Surveys asking existing users about their on-site significance.
  • Getting expert opinions on your site’s design (heuristics).

When you use a website often, it’s easy to become snow blind to it and overlook its faults. These research methods will allow you to see through this and determine how new users might react to a site. This is crucial if you are trying to decrease bounce rate.

CRO planning and testing

The methods above will allow you to generate plenty of data about how people use your site. Once you’ve got that, the next step is to use it to see how you can remove any obstacles to conversion.

Key questions to ask of your data to determine what these obstacles include:

  • Is your website allowing users to find the information they’re looking for?
  • Are people getting distracted while on your site?
  • Are the important elements of your site getting noticed enough?
  • Is your site targeting the right people?
  • Are your calls to action (CTAs) effective?
  • Does your site look trustworthy to users?

Once you’ve answered these questions, the next step is to develop your CRO plan. This will consist of strategies designed to bypass any problems.

You’ll often find it difficult to choose between CRO strategies or approaches. This is where testing comes in – which will help you to determine the best method to use. Testing can often increase conversion rate in a big way – so it’s an important step to take.

Testing in CRO takes two main forms – split testing, and multivariate testing.

  • Split testing tests two or more versions of a whole webpage. These versions get pitted against each other to see which is the most effective.
  • Multivariate testing involves testing webpages created from a matrix of different elements. Multivariate testing can get complicated, but guarantees you will find an effective solution. It works best with high-traffic websites.

It’s important to note that even once you select a CRO solution, you must keep tracking your site’s conversion rate. This will allow you to see if any changes take place which demand further optimisation. These changes can often be due to factors outside of your control.

Structure and repeatability are both important factors when putting CRO into practice.

Companies with a structured CRO process are more likely to raise their CRO budgets year on year. This implies that what they are doing is a success – otherwise why invest more money in it?

But 65.7% of companies have no documented CRO process. This means that 65.7% of companies are missing out.

With this in mind, best practice is important when planning any CRO research. Tips here include:

  • Have a dedicated person / team who is accountable for all CRO within your company.
  • Share knowledge of test results among relevant personnel.
  • Archive test results to avoid duplicated effort.
  • Ensuring you let tests run until you have achieved a good sample size.

Conclusion: scientific CRO is effective CRO

What we’ve shown here is that CRO itself is no safeguard against a poorly-designed website. You must do CRO the right way – which is always the scientific way. Scientific CRO is exemplified by the following factors:

  • Research.
  • Planning.
  • Testing.
  • Repeatability.

All the CRO you do should follow this pattern. It’s also important to document your processes and make someone responsible for your CRO. This would include curating official CRO guidelines and training others in their use.

Scientific CRO will ensure that you direct your efforts in the right way. It will also ensure that they are not duplicated, and that they follow a common structure. In larger organisations, where several teams may be working on CRO, this is invaluable.

A fine line exists between useful guidelines and pointless bureaucracy, but this is the line that CROs tread. Effective training will help to ensure that systems are a help and not a hinderance.

The next article in this series takes a look at exactly how CRO is done. What are the magic techniques that make users click on your buttons?

But first, we’d like to know a bit about you. How do you plan your CRO? Do you have a set of guidelines or do you follow more of a scattergun approach? Let us know in the comments section below:


Author Matt

More posts by Matt

Leave a Reply

Enjoyed what you read?
Join to receive more great content delivered to your inbox.
Thanks for signing up
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nullam convallis nisl dui, id tincidunt velit maximus eu. Praesent viverra rhoncus aliquam.