Last updated: 9th November 2018
Like most disciplines in business – there’s three things required to make it work:
There’s a plethora of A/B testing and personalization tools and software solutions out there – the most common in Australia at the enterprise level tend to be Adobe Target, Optimizely, Oracle Maxymiser, Google Optimize and VWO.
You need to make sure you're using the right one for your requirements (which may include developer friendliness, security, enterprise-level features like changelogs, etc) -- and the journey begins there.
As expressed so succinctly by Gartner in 'Magic Quadrant for Personalization Engines', 'most organisations are very willing to invest in the right marketing technology solutions at a hefty price tag -- but less willing to invest in the people with the right skills and knowledge to drive adoption and success.
Who's going to do all of this? How many hours are required? Across what roles? If you're planning to use internal staff members in-house -- can you verify that they have the right skills? Are they very busy right now?
There's a lot of things that can go wrong when you're working in a fast paced conversion optimisation program -- tests not being QA'd properly, tests not meeting your source code standards or causing problems/flickering/unexpected user problems, audiences and metrics being misconfigured -- the list goes on.
The best way to solve these problems is with solid well documented policies, procedures and governance surrounding your CRO program. In future blog posts, we will share a number of our processes that you can 'copy and paste' into your program for better efficiency.
Conversion research is, and should always be, your starting point (once you have the right execution platform, team and processes in place). If you wish to beat the ‘industry average win rate’, conversion research is essential.
There are a variety of conversion research approaches, and ‘conversion research’ may have a strong overlap with many of the activities already undertaken by your VoC (voice of customer) team or UX research team. Within an enterprise, this means you should tap into leveraging insights and finding out what has already been completed in other departments throughout your organization.
All web sites, brochures and pieces of marketing material have a natural “visual hierarchy”. Careful use of design (colours, layout, typography, etc) affects the way (and order) in which the viewer of the content absorbs and reads the information.
Computer-based mouse tracking software is one way you can quickly determine which components of your design are the most ‘attention grabbing’.
For example – what components of your design/layout stick out? Is the main ‘CTA’ easily missed (perhaps it does not look like a button or does not contrast with the colours used in the core design?).
By utilizing a tool like EyeQuant (which admittedly comes with a hefty enterprise price tag) – it is possible to quickly and effectively get a quick snapshot of what components of your design stand out.
It’s important to note that there is only a high correlation between what tools like EyeQuant are able to uncover versus true (more expensive) biometric lab testing and formal eye tracking.
Checking which components of your design ‘stick’ out the most allows you to identify conversion problems (i.e.: no one is noticing the CTA or no one is reading the persuasive headline). Once you have problems – you can begin to brainstorm a number of possible solutions (‘conversion treatments’) for each identified conversion problem. Each of these should be run as a variation in an A/B test versus the control problem (as explained in other blog articles, we highly recommend running multiple variations per test to dramatically increase your win rate).
Visual hierarchy and click or (more advanced) eye tracking is an important part of utilizing biometric-style signals to get to understand how your customers actually feel, think, breath and see while experiencing your online customer journey.
You don’t want to only focus on your website – another fantastic source of potential test ideas can be your competitors’ websites. This can help you evaluate current trends and formulate new ideas.
That said, it doesn’t mean that you need to copy everything that competitors are doing as they may not be doing everything right. Some industry verticals suffer from over-focusing on their direct competitors – and fail to come up with new innovative ‘out of the box’ ideas (that nobody else is doing).
If you see a new tweak or change that your competitor has rolled out – you should test a similar addition to your site to see the effect it has (don’t assume just because your competitors are doing it – that it works).
It is best to evaluate what’s relevant and suitable for your customers before implementing any new ideas. At the very least, reviewing your competitors’ websites can help to generate ideas that you can then consider testing.
While analysing a website (or mobile application) at eyeshot is the vital first step to evaluate layout, copywriting and psychology, it is not enough. Your web analytics data (used correctly) can be a powerful source of identifying potential conversion problems and ‘finding out what’s going on behind the scenes’ (by investigating the raw data).
These days and at the enterprise level, there’s a plethora of analytics data available. Most businesses have one or many web analytics tools at their disposal – most commonly Google Analytics or Adobe Analytics.
By probing into your web analytics data, you should probe for answers to a variety questions. Here’s a few common ones to get started:
A well experienced conversion rate optimisation professional or web analytics specialist in your team and/or in your partner agencies can help dramatically in terms of finding answers to these questions quickly and efficiently – but really, you should expect most of your digital marketing team to be able to jump into your web analytics tool of preference and quickly find answers to at least the above questions.
One of the major frustrations with using web analytics data for ideation is that the data may often show you that something is broken, but not how to fix it. It can also sometimes lead to meaningless false positives.
Let’s imagine a hypothetical scenario in which a client has an analytics system that reveals mobile conversion rate is 89% lower than desktop conversion rate. Does this mean that your mobile website is suboptimal? Perhaps. If that is the case, what changes do you need to test and hopefully apply to improve mobile website experience? This can vary dramatically.
Does it always mean that your mobile website experience is broken simply because the mobile conversion rate is lower than the desktop conversion rate? Probably not – perhaps, your website visitors prefer to take action on desktop – not mobile. (Mobile conversion rate tends to be, on average, a lot lower than desktop conversion rate across most industries regardless).
Ultimately, data and facts from your web analytics system can help to reveal insights which lead to a better CRO program with a better win rate, and better ideas.
Heat maps are simple tools that can provide an abundance of useful data and meaningful insights. They are capable of displaying complex data in a simple visual format that is easily understood.
We once conducted a split test for a company – there was a different variation for both the mobile and desktop device – with the results revealing that the mobile version won whilst the desktop version yielded a slight loss.
An anomaly was noticed in the desktop version. 5% of the desktop visitors were clicking on the business logo in the upper left-hand corner, while a staggering 50% of the mobile visitors were clicking on this same logo on the mobile version. Why would this be?
After scouring the page over to understand why this was happening, it was concluded that the mobile version was missing the live chat function. The mobile visitors were clicking on the logo as they were seeking help.
This is where heat maps come in handy as they provide supporting evidence to conclude why an unusual anomaly might be occurring. Heat maps also help in determining the factors of a losing test.
Visitor sessions show you the exact cursor movements of your customers through a video. The video records the cursor movements of your customers, their scroll patterns, what they click on, etc.
Our findings revealed an interesting fact: Website visitors’ behavioural patterns differ within the day and time.
We once analysed the visitor sessions for a healthy chocolate chip cookie website. We noticed that their customers generally scroll through the site methodically during the day and night but interestingly, their customers zipped through the site in seconds without any purchase in the early morning hours.
This behaviour could possibly be explained by cravings for sweet treats occur at a higher rate from mid-day as compared to mornings.
Such visitor sessions may also provide meaningful insights on how your customers interact with your website during weekdays as opposed to weekends.
User tests are a valuable way to assess the challenges encountered by your customers and have these challenges documented clearly.
Usability expert Jakob Nielsen states that 5 user tests are suffice to reveal 85% of your website problems. In other words, you need not stretch your marketing budget on 100 user tests as it’s likely that you will just be collecting repeated information.
Most web hosting services provide a web statistics platform and it is essential to have one in place.
Google Analytics is considered the holy grail of web statistics. While most of the data within is self-explanatory, there are also methods for filtering and extracting the most intricate data.
If Google Analytics is over your head, you may find it well worth the investment to hire a certified expert to sort out the data.
Surveys are another hidden gem that is often underrated. It can be tricky to sort through all the responses especially if there are open ended answers.
This is where surveys prevail in areas where other forms of analytics fail. Surveys are a platform to collect your customers’ thoughts on any questions you have in mind.
We once worked with a client who offered a relatively attractive value for money print coupon that greatly discounted their products. Strangely, the sale conversions didn’t happen. We ran a survey with his customers and interestingly, the results revealed that this marketing campaign failed as many of his customers didn’t own a printer and therefore couldn’t utilize the coupons.
Do your customers immediately know and understand your offering? It is critical that a website conveys what they have to offer within 5 seconds of landing on the page - here’s why:
Your customers are very busy people. If they were to search 10 different sites for 5 minutes each to try to determine if they had what they needed – they would have wasted almost a full hour of their time and very likely end up feeling frustrated;
If you tell them immediately, right up front, what you have to offer, they’re likely to stick around on your website. Clearly, conveying your offering upfront should lower your bounce rate.
Once you have addressed the “5 Second Test”, review the website to see if it adheres to the Web Marketing ROI 10 core conversion principles and that it provides the below elements to support those principles. By complying with these principles, you can take away prospect’s anxiety and be a step closer to converting their “wish list” into purchases.
The ten core conversion principles work because they map to the overarching principles behind why certain designs, websites and mobile apps convert better than others – in fact, these principles have been in marketing since the days of direct response and direct mail copywriting – long before the Internet ever was invented.
Here’s a brief explanation of how just a few of the 10 Core Conversion Principles work and how to use them. (In our experience, getting good at using a conversion methodology like this for ideation often takes years of practice, but it’s a powerful model if used correctly).
Brand Awareness is the newest addition to our ‘core conversion principles’ methodology at Web Marketing ROI. How do you explain that some brands (who are trusted long before a website visitor arrives at the website) naturally convert much better, sometimes even with a sub-optimal check-out experience? Brand.
Brand can be summarized as the attitudes of the consumer to your website and brand, before they actually arrive and visit. For example – you may naturally trust and have heard of Nike or Chemist Warehouse (preferring them because of name recognition, offline TV advertisements, and so on – over say, a smaller lesser known pharmacy).
Once you’ve clearly conveyed what you have to offer, your customers will at the very least step back from immediately bouncing. But your initial work isn’t done just yet.
They now know you have what they need, but they’re probably asking themselves “Why should I choose you?”
They may feel some anxiety surrounding there being a better choice out there and are afraid to make the wrong decision. It is not unusual that your website visitors wish to avoid experiencing buyer’s remorse. You can alleviate their fears by providing an immediate reason why they should choose you over your competitors.
Do not assume your customers will trust you immediately. There are thousands of business websites out there and online consumers today are very wary of scam websites.
Provide trust indicators such as security seals, testimonials and privacy reassurances to build their confidence in your brand and business.
We’ve come across many examples of website copy that is both confusing and unclear. It is difficult to move forth with a transaction when your customer is confused. Keeping your copy clear and concise is the key to clarity.
It is critical that your customers receive messages that are relevant to them. Irrelevant messaging may result in high bounce rates or your customers dropping off from your website altogether.
A purely factual approach may not be the best marketing strategy. Find interesting ways to ramp up the desire and perceived value of your service offerings to differentiate yourself from your competitors.
For example, if you offer newsletter subscriptions, make sure you clearly convey its value and privileges that only registered account users are entitled to. This way, your customers know what they are signing up for and want to be in your mailing list.
This is one of the most common problem encountered by businesses. Several businesses unknowingly create obstacles that prevent their customers from making a transaction successfully.
Many websites make it nearly impossible for their customers to interact with them due to problems such as slow page loading speed. Slow page loading speed not only affects conversions but SEO rankings as well. The Google algorithm tends to favour websites with faster page loading speed.
This is an important element that most businesses do not utilize enough. Having your customers feel that they have to purchase your products and services immediately is a very powerful tool. Allowing them to feel safe in putting off purchases for a few days, or even a few hours may kill conversions.
So there you have it. Work through all of the above - start with the critical conversion research methods, then the “5 Second Test”, use the “Core Conversion Principles” to improve your enumeration of conversion problems/multiple conversion treatments per identified conversion problem and be sure to dig deep into analytics – to add value to your conversion optimisation process.
It’s a strategy that’s bound to increase your conversions and your bottom line.