Website conversion rate optimization is truly one of the current buzzwords in digital marketing. What began as a phrase whispered among online marketers has found its way onto the tongues of multinational CEOs and amateur entrepreneurs around the world.
Now, everyone is raring to get stuck into CRO, but many dive into unnecessary testing that fails to yield actionable results. I’ve already made this mistake for you, so here’s a compiled some essential conversion rate optimization tips that DO work – to help you ensure that your efforts pay off.
Don’t run before you can walk
Our society encourages us to dream big, which can be great. I’m certainly not against ambition for success, but it is important that we don’t get carried away before considering the availability of resources.
In an ideal world we would set up landing page testing for every possible avenue on our site, and split test every possible variable, from calls to action to what color underwear we have on. Left to their own devices, clients often over invest in the planning stage of CRO and end up with hugely complex testing plans. In theory all this structure is great, but the reality can be overwhelming.
If you are new to CRO, start off with a single landing page test. This will give you a basic understanding of the process from start to finish, and will help you plan more complex testing in the future. Dream big, but start small.
Understand your traffic: How much? From where?
Conversion optimization testing is only effective when your results are statistically significant. This means that you can only conclusively say that variation A is better than B when enough data has been collected. The math is relatively simple, but there are plenty of calculators out there to make it effortless. Many A/B testing tools will calculate this for you.
If your site has a low volume of traffic, it may not be the right time for you to carry out conversion rate optimization. It would make sense to focus on online growth strategies to build up the requisite data required to carry out tests. If this is the case for your business, stick to best practice for now, and get in touch for some innovative ways to acquire new traffic.
For those with a steady flow of visitors primed for testing, it is important to consider the source of the traffic. Users arriving on your site will have different intentions, depending on which keywords they used, or which social media platform they clicked through from. Consider how strong the purchase intention of each user is, and ask how you could segment and treat them differently.
For example, a user coming through Adwords using a strong purchase intent keyword may respond differently to a user who clicked through your company Facebook page. Therefore, consider your channels, and tailor your landing pages and testing structures to each segment.
Beware of multivariate testing
Similar to the previous points about traffic volume and complexity, multivariate testing will only yield results for those with a very high volume of traffic. This may be an effective testing method for monster e-commerce sites like Zappos, but most site owners will just waste time and resources on tests that would take years to deliver actionable results.
Stick to A/B testing unless your traffic volume is high. There will be less ambiguity about your results, and you can compound your learnings into other tests to inspire a more methodical hypothesis for other pages. Most importantly, you will get feedback from your changes quickly.
Setup conversion tracking
It is impossible to optimize your site for conversions without a proper tracking infrastructure in place. For most sites, this encompasses a correct installation of Google Analytics, preferably using Google Tag Manager for easy event tracking. These two tools should allow you to track almost anything on your site, and will help you gain insight into your page performance to understand where to run your tests.
There are also a variety of testing tools out there, from Optimizely and Visual Website Optimizer, to more expensive solutions like Adobe Test & Target. These tools all have their pros and cons, so research thoroughly to pick the one that’s right for you.
If the goal of your site is to collect telephone leads, you must set up some kind of phone tracking. Unfortunately, most free solutions for this will only track those directly clicking on the number from a mobile device, but there are other affordable solutions to track calls. These technologies combine real time session data with rotating numbers to identify individual users. The reality is – if your conversion goal is a phone call, you will not be able to do conversion optimization without call tracking.
Have a clear and compelling call to action
Just about every conversion rate optimization blog in cyberspace will proclaim the importance of the call to action (CTA). This is essentially juiced up marketing jargon for the use of a verb in imperative form (eg. Shop now, buy this, click me, shampoo your beard).
Having a CTA is certainly important, as it plants the seed of an idea in your user’s head to follow a particular action. Shampooing your beard right now? No, it is not mind control, but it does help to steer the path of the user towards your desired goal.
Many site owners read about the all important CTA, and pepper their page with them. Identify one primary goal, and make that consistently the focus throughout the page. If you have other calls to action (eg. Read more on a blog article), ensure that your primary CTA is bolder and more visually appealing.
On the one hand it is important to keep your command simple, but don’t be afraid to escape the generic. If you have a very specific action, then be specific in your CTA. Experimenting with the wording, and design of your CTA is a high impact A/B test for your site.
Those who dare, win – so try out something a little outrageous
Best practices are great guidelines, but wouldn’t the internet be a boring place if everyone just stuck to the same thing? By trying out more radical copy and design elements, you are testing the boundaries of your users. Learning from their reactions to “out-of-the-box” ideas will give you a far deeper understanding of their mentality than split testing two more conservative designs.
Start with a hypothesis. If you believe that your audience is risk averse, and a little bit prudish, then tastefully test them with something a little wilder. Similarly, if you feel married to a particular page layout, try out something completely different to challenge your assumption.
If you are worried about causing some collateral damage to your brand, then target the test at a limited traffic segment to gauge a response first. In my experience, companies are over-cautious about such issues, which is understandable as a brand is built up through hard work and dedication. However, remember that you are not nurturing your brand if you allow it to become stale and ordinary. Take some risks to reap bigger rewards.