Our Unique Conversion Rate Optimization Methodology

ThinkTraffic has been a valuable partner with their one of a kind off-site optimization services. Great results within a reasonable amount of time!

Adam Jankovits

The idea behind conversion rate optimization is pretty simple, but that doesn’t mean that the process is simple. Successful split testing requires a bit of art and a lot of science, carefully implemented to ensure the best results.

Here’s an outline of our own methodology which we use as the structure for each of our CRO projects. This method ensures that our time is spend on the things that are likely to produce the best results in the shortest time.

Step 1 – Getting To Know Your Business

A website can have any number of important metrics and some matter more than others. Of course revenue is what really counts, but for practical reasons we need to understand which metrics drive revenue and what data best represents the effectiveness of your website.

The first step is to establish what our goals will be for the project and if possible to find a benchmark based on how your website is currently performing.

Typical primary goals might be:

  • Contact form completions
  • Leads generated
  • Checkout completions
  • Products added to cart

If you choose the wrong metrics it is entirely possible to improve conversion rates without really impacting revenue or profits – skip this step at your peril!

Step 2 – Understanding Your Traffic

Here’s where it gets interesting. At this stage, many companies simply guess what factors might be effecting their conversion rates, but that’s a good way to get nowhere fast.

Instead, we analyse your traffic and how it is currently behaving. We use a handful of tools and techniques as required:

  • Traffic flow analysis
  • Primary traffic sources
  • Mouse attention maps
  • Click attention maps
  • Session recording

All of which helps us to better understand how visitors are using your site, which parts of your site are underperforming and where the biggest wins can be found.

Not every split test results in an improvement, but by doing the leg work we can assure that when a winning combination is found, the improvement is substantial enough to make it worth the work.

Step 3 – Understanding Your Competition

Of course, your business doesn’t exist in a vacuum. There are other alternatives out there and even if your product is fairly unique there will be similar or related products to consider.

So before we start designing your first tests, we will look at what the competition is doing. If they are being successful then we may be able to take ideas from their site design or how they are differentiating themselves.

Step 4 – Designing The Tests

Once we understand the project we can begin designing the split tests. By this point we will already have a few ideas for things that we would like to test and by using our experience and research we can prioritise and test the items which we think are likely to have the most profound impact on conversion rates.

Exactly how we implement these tests will depend on what we want to test, how your site is currently performing and how much traffic is available.

By testing big changes first we can usually start drawing conclusions relatively quickly…

Step 5 – Drawing Conclusions & Next Steps

Conversion rate optimization should be an ongoing effort. Each test can have 2 possible outcomes:

1 – The new design loses

The new design “loses” if it fails to result in a better conversion rate than the existing version. This does happen, but it’s not really a failure for a couple of reasons:

  • It’s a numbers game; you only need a single big win to make all of the losing tests worth the effort.
  • If the new design fails to beat the old design, then you keep the old design and move on to the next test.
  • Every test provides new information which we can use to inform our future tests.

2 – The new design wins

Of course, this is what we want to happen, and frequently it does happen. If we discover that the new design outperforms the old version to a statistically reliable degree, then we will keep the new design.

  • The new design becomes the control
  • We create a new, new design and continue testing

The reason that this technique is so powerful is that if the new design loses, we scrap it straight away, but if it wins – your business gets that conversion rate improvement every month from now on…

Whatever the result, we draw conclusions based on those results and we design the next competitor to be tested against the new control, whether that be the victor from a previous test or the original.

Incremental Improvements Add Up

The really powerful thing about this methodology is that we are always testing the current best design against new contenders. So let’s say that we start of with version A of a page (the original):

  • Version A converts at 5%
  • We test version A against our design (version B)
  • Version A wins, so we keep it and then design a new test (version C)
  • Version C wins with a conversion rate of 5.4% so we keep it
  • We run a couple more tests (versions D and E) and C wins
  • In the next test, version F wins with a conversion rate of 6.1%
  • That’s a total improvement of over 20% (5% -> 6.1%)

As you can see, each win brings a slightly better conversion rate, so by the time you have run 3 or 4 successful tests you might have increased your conversion rates by a substantial margin.

What would it mean to your business to increase your sales and revenue by 20%, 30%, 50%?

Get In Touch Now & We’ll Show You How!

Your Name:
Email Address:
Your Message: