Conversion Rate Optimization, otherwise known as CRO, is an incredibly efficient testing process that can enhance the performance of your website.
If you’ve got a good level of traffic to your site but visitors simply aren’t biting, CRO can help you to identify and overcome conversion hurdles in your sales funnel.
By optimising your web conversion rate you get more people to take action – without having to spend money attracting more visits to your website.
Is Conversion Rate Testing for me?
It is often difficult to justify spend on a relatively new and unknown marketing technique. Yet, CRO can get concrete results quickly, improving return on investment (ROI) in a relatively short time – often within the first three months of the project. The larger your web traffic and turnover, the bigger and better the improvement is likely to be.
Some of our clients have seen incredible improvements in conversions, decreased bounce rates and more sign ups simply by testing different variations and combinations of their site’s layout and content. Let’s look at real tests we ran for one of our clients, an online retailer that sells luxury items.
What’s the process?
Before we set up tests for our clients, we get to know their business. We identify what’s leading visitors to convert and what’s causing visitors to drop out of their conversion process. With CRO, it’s important to know what the problems are, so your tests are targeting specific issues, rather than testing to find issues.
Once we have identified conversion problems, we’ll create experiments to see if we can overcome them. So, for example, many eCommerce websites suffer from high bounce rates. This could be because a product page is too cluttered and the product benefits aren’t obvious enough. We might then test different page layouts to see if we can make the products’ benefits more obvious.
We’d then think of more effective ways to communicate benefits on the product page(s). Using these different hypotheses (or research questions, if you think hypotheses sounds too posh), we’d then experiment and run split tests on their site. These might include variables like a bigger headline, higher resolution images of their products or even a different colour(s) for the “call to action” button.
Once we’ve gathered enough data to formulate concrete conclusions, we let the customer know what’s worked. They’ll have actionable data to roll out improvements across the whole site, leading to greater conversions.
However, CRO is an ongoing process and does require further investments of time, resources and testing tools. This exercise is something you should continue to do and refine over time in order to make sure sales processes are always in good shape.
To give you a flavour of the impact CRO could potentially have on your business, I’ve picked a few tests from our recent CRO projects. Each numbered section highlights the research questions our experiments investigated.
1. Present a clearer template layout, remove clutter and show what matters to the visitors.
This particular client’s eCommerce website suffered from high bounce rates.
We know that the majority of the conversions, if not all, took place on the product pages. The quicker we could get visitors to their desired product pages, the more likely they would be to convert.
Reviewing the category page template, we realised the miniature “Free Shipping” banner below the product image was repeatedly displayed on all the category pages. Although it’s a great offering, unnecessary repetition of the same message in a small space can be disruptive. Here’s what it looked like:
The client was also displaying an “In Stock” banner (see above) for products that were invariably in stock, but it wasn’t necessary at all because a customer’s impression is that all the products should be in stock. If some aren’t, instead of initially telling them the product isn’t in stock, causing them to bounce off the category page straight away, we could take them to the product page and capture their information through a “subscribe to notification” function to receive an update when the product is back in stock and cross sell if there are similar products.
What test did we run?
With this hypothesis in mind, we decided to create an A/B test, where the variation didn’t have the “In Stock Banner”. A/B testing (sometimes called split testing) comprises two or more versions of a web page to see which one performs better. You compare proposed variations with a control version by showing them to similar visitors at the same time. The one that gives a better conversion rate, wins!
Within four weeks, the test generated a conclusive result with the below variation out-performing the control version on the bounce rate by 28%.
At the end of the test, I did a bit more research on the internet and found that many other eCommerce sites have stopped displaying “In Stock”, “Sale” and even “Add to Cart” banners. This, I believe, works well for big ticket items, however for the impulse buy (small price) items; it may have a complete opposite effect.
Below are two examples:
As you can see, these fashion items are priced higher than the majority of goods on A.S.O.S. All three are displayed without “Add to Cart” banners.
On the current product page for Ebuyer.com, you can see that the add to basket options have been dropped.
2. Tell visitors directly why they should buy from you. Don’t ask them calculate a discount!
Product pages are your money pages; this is where the majority of the conversions take place for eCommerce sites. That’s why optimising them are high up on our list of priorities.
Our aim is always to make the visitors’ life easier by providing the most important information upfront. We achieve this by removing any unnecessary steps required of them. This should improve the shopping experience, and thus the conversion rate.
What test did we run?
Price is the most important information on a product page in many cases – it certainly was in our client’s case. They even offered a 110% price protection.
However, we noticed that although the current price and retail recommended price were both provided, the amount saved wasn’t explicit – meaning visitors had to calculate it themselves. So we decided to run an A/B test with the variable displaying the saving. Below you’ll see the control and the variation – the winning variation produced a 38% reduction in bounce rate over a month’s period of testing.
3. A picture is worth a thousand words
Pictures speak louder and clearer than words – and if you want the quality of the goods to speak for themselves it’s a good idea to incorporate high resolution images on your product pages.
The improvement of broadband speed over the past few years has meant more consumers can handle multimedia intensive webpages without crashing. YouTube’s popularity underlines this evolution perfectly.
Beautiful and higher quality imagery may require a slight compromise in the page load speed, but this will be more than made up by the awesome impression images will have on visitors – provided that the page load speed remains reasonably fast.
What test did we run?
To put our theory to the test, we created a testing variation with a product zoom-in window enlarged by up to four times more than the control version – the original version.
And the result has surpassed even our wildest expectations. Our eCommerce client saw a 221% increase for the number of clicks on the “Add to Cart” button on individual product pages!
4. Colour, colour, all important colours!
Legend has it that Google once ran an A/B test for hundreds of shades of link colours in its SERPs to improve click through rates. Imagine if they produced only a 0.001% improvement; that could translate into tens if not hundreds of millions of pounds of revenue for Google because of the sheer traffic volume it receives.
It’s one of the simplest CRO tests in terms of the set up, and yet the results can be most surprising. The idea is that it should get the visitors’ attention, but the colours shouldn’t be too invasive or offensive. A good contrast compliments the colour scheme of the website and should be an appropriate colour to reproduce on different screens.
What test did we run?
We wanted to find the best colour for our eCommerce client’s “call to action” buttons on their product pages. To do so, we selected seven different colours that complement the main colour scheme. We ran the A/B test and purple won by an incredible margin: a 344% increase on the number of clicks on “Add to Cart” button!
Here are our seven colour variations, plus the control, below:
Finally, these findings may or may not be applicable for your business, however the process and ideas are reusable and the more traffic your website receives, the better outcome these tests are likely to produce. We hope these case studies will give you fire power in the boardroom, after all, it’s ROI that speaks louder than anything, and CRO can produce verifiable and reproducible results.
Get in touch
If you are interested in running conversion rate optimisation tests on your site, get in touch using the form below and we can discuss the most appropriate tests for your business.