In business as much as in life, change is hardly negotiable. The only thing you can do is try to gain some control over HOW that change plays out, but not IF it plays out. When it comes to your website and its conversion rate, it is crucial for you to understand that leaving things as they are will eventually work against you. Maybe not this week, maybe not next month, but likely so in the not-too-distant future.
Yet from our own experiences, we often see conflicting attitudes towards CRO: some people cannot get enough of AB/ABn testing, whilst others only think of it as an unnecessary hurdle. The latter group tends to shy away from using CRO in their decision-making process. Yet sometimes with CRO, it’s a case of JFDI - just effing do it! If you take the path of no action, you might end up in more trouble than if you had done anything at all. And if you do take action but it’s the wrong action, you’ll most definitely be in trouble.
Conversion rate decreases over time
Unless you are a giant in your particular business vertical (e.g.: Amazon), visitors are not going to be impressed if your website looks the same year after year. Ecommerce is in constant evolution alongside society itself. 10 years ago, mobile devices did not commonly surpass desktops on the conversion rate metric. Now they very often do, in no small part due to the fact website designers have adopted a mobile-first attitude, which has massively improved users’ experience on the smaller screen. And even if people also evolved to be more reliant on mobiles to start with, an outdated mobile look & feel might still repel the majority and hinder conversion.
Your competitors understand all of this, and they will be dedicating time and resources to make sure they have the edge on you. Whilst benchmarking against competition is not always a foolproof way of informing decisions, if competing websites renew themselves more often than yours does, you run the risk of falling behind on the (perceived) innovation track. The last thing you want is for a visitor to think your website does not look as enticing as your competitors’.
For the reasons mentioned above (and many others), websites have a natural tendency to see their conversion rate decrease with time if nothing is done to arrest that decrease. Businesses that fail to change eventually run the risk of going out of business altogether.
Change is not inherently positive
Depending on how you execute changes on your website, the results might not always be positive. It’s not because something is newer that it’s better; it’s got to be the right thing! We generally see two different approaches when websites changes are carried out:
- The one we like to call DAB DAP (Design & Build, Deploy & Pray!). This is a before/after approach, which we often see in bigger changes such as whole website redesigns. In this case, your update is deployed with the assumption it will have a positive impact.
- The one where you keep a Control group after deploying your update. The assumption is that what you think is going to have a positive impact could be proven wrong by your website visitors.
Obviously, the DAB DAP approach carries a much more significant risk compared with the Control approach. In comes CRO…
CRO as a risk mitigation tool
Especially for audiences who still aren’t convinced about the benefits of CRO, this form of testing should be pitched as a very effective risk mitigation tool. Matt Scaysbrook recently blogged about whole website redesigns often being used as a tool to solve ailing conversion rates. But more often than not, they do the exact opposite. The reason why this frequently happens is not rocket science either: big changes = big gamble!
By providing a Control vs Variant approach, AB testing is a sane way to ensure that whatever changes you have made are not only delivered in a measured way (e.g.: you can throttle the test’s traffic level), but they are also totally under your control (you can stop the test whenever you want). If the new design doesn’t work, CRO testing provides a kill switch and an easy way to roll back to the Control version that worked better. Some CRO platforms even have built-in features allowing for the automatic removal of losing variations whilst the test is still running. This comes in handy in the context of ABn tests with multiple Variants being tested against a Control.
CRO is an informed way to make changes
If your test results are negative, using CRO as a tool also allows you to take the time to reflect upon the reasons for the failure and make the necessary adjustments, ready for a retest. This is what we call iterative testing, whereby you build upon a previous test in a bid to redress a metric that could have yielded a better result.
Unlike the DAB DAP approach described earlier, CRO testing allows you to dig deep in your analysis of the results, all the more since the data is more objectively comparable than with a before/after approach that could be influenced by external factors (e.g.: seasonality, marketing campaigns, etc.).
CRO testing allows you to ask the right questions:
- Was it a particular audience type that didn’t respond to your new design (e.g.: Paid traffic, Returning users, etc.)?
- Was it a particular place in the user’s journey (e.g.: PDP, Basket, etc.)?
- Was it a specific feature that lost traction with your visitors? (e.g.: Interaction on key CTAs, etc.)
CRO shouldn’t be optional
CRO testing deals in facts and data, using statistical significance as its stamp of objective validation. It’s basically as scientific a way as can be of making necessary changes. You shouldn’t be afraid of making the changes you want to make, but don’t just release them unless you’re ready to accept the fact that you’re taking a big risk with your visitors’ reactions.
Even if CRO does require an investment in terms of knowledge, time and money, resources spent on testing seldom go wasted. In conclusion, CRO testing should not be optional in any business, it’s a JFDI situation where you have more to gain than lose from using it and carrying out changes the right way.
.png)
