Oxford College of Marketing tutor, Alex Hutson, looks at the top 5 tools organisations need for successful usability testing and conversion rate optimisation (CRO).
Ideally, all organisations, whatever their purpose, should be focused on satisfying their customers’ needs and expectations throughout the customer or user journey.
If you don’t work in a B2B or B2C context, then consider your ‘customers’ as the people that it’s your organisation’s purpose to serve. For example, before I worked in marketing I worked for HMRC and our customers there were the UK tax payers, and when I worked at The Freemasons’ Grand Charity our customers were essentially the members (the Freemasons) who donated millions of pounds every year to other charities.
So before reading further, just make it clear in your mind who your customers are, and that it’s highly likely that there may be more than one group or segment of customers. Do this, and you will be able to learn more from this blog post.
I mentioned ‘users’ at the beginning of this post, that is because in an online sense your organisation’s interactions are with ‘users’. Some of those users will be potential customers and some of them will be current or lost customers. And you must remember that it’s more than likely that their customer or user journey takes place through a variety of online and offline interactions, using more than one device (e.g PC, tablet and mobile phone) and on several different websites and platforms (e.g. search engines, social media channels, etc.).
Typically, an organisation’s digital presence has a purpose to either provide one or more services (e.g. HMRC: submit a tax return through the portal; TalkTalk: to provide customer account information, facilitate service changes, provide billing info) or generate leads and/or sales.
When these actions are taken, they can me considered as macro and micro outcomes or conversions.
To be honest, I’m a bit obsessed with making marketing accountable. To this, we need to be able to measure and test everything. In my opinion, Avinash Kaushik’s blog is the best in the world for tips and guides on how to do this.
If users are having a good experience along this journey they will be more likely to find what they are looking for and like what they find. In turn, this means that they will be more likely to take the desired actions (the micro and macro conversions) and therefore conversion rates will increase. We can see evidence of this from the MarketingSherpa Ecommerce Benchmark Study companies that were reactive (only being concerned about ease of use when there was a problem) were less likely to meet revenue goals, with 40% of these marketers falling into the “slower than forecast revenue growth” group.
How often do you evaluate the following aspects of your web pages or templates?
So usability testing and optimisation (making it better) should form part of an organisation’s conversion rate optimisation (CRO) work.
Fortunately, for us marketers, there are now lots of tools available that can help us do this.
The following usability testing and CRO tools are my five favourites for discovering what users are doing, why they are doing it and for testing improvements.
These three things are all necessary in order to optimise the user experience (UX) and the conversion rates.
I use all of these usability testing tools on a daily basis and I recommend them for providing excellent insights and great value for money.
#1 Google Analytics
You can use Google Analytics for free, which is amazing!
This tool is so powerful, and when it’s set-up correctly the reports that can be generated offer fantastic insights.
From a usability testing point of view, its main use is to help us see what is going on and where improvements could be made.
When working in Google Analytics I use many different built-in and custom reports. For this article, I have covered three key reports that can help you gain usability related insights:
The Behaviour Flow report (pictured below) shows how people move around a website (or app) from page to page.
The focus is on the most commonly used paths but you can drill down into higher levels of detail.
The default start of the report is based on the Landing Page, but you can change this. For example, if you select Source / Medium then you can see where users have come from before they start their online journey. This is great for starting to establish the intent of the user and how that affects the journey.
Look out for the following:
- Users going back and forth between pages as an indicator of confusing information
- Visits to unexpected pages as part of the journey as an indication of misleading navigation options
- Large volumes of users exiting from certain pages as an indication that they are frustrated or unhappy with the page content or that they have managed to complete the task they set out to do on the website
- Early significant drop-offs relating to entries from a particular Source / Medium as an indication of misleading off-site information used to generate visitors
The Events Flow report (pictured below) shows event sequences in much the same way as the Behaviour Flow report.
This is important because not all actions on a website revolve around viewing different pages, they involve actions being taken on a page (e.g. submitting a contact form, using the live chat box, watching a video, downloading a lead generating PDF, etc.).
Again, the focus is on the most commonly triggered event paths but you can drill down into higher levels of detail.
You will require someone with technical expertise to set-up your website or app so that user actions trigger website events. A great tool for doing this is Google Tag Manager.
In the Behaviour section of Google Analytics, you can access the Site Search report. If you have a search function built into your website then the results of the searches are shown here.
Basically, if users can easily find what they are looking for on a website then they don’t need to use the search function. So where you have searches there is an indication that your navigation system is confusing or misleading.
Then if you look at the searches performed, this is then a more specific indication of what content users are struggling to find.
If several users are searching for the same things you can spot themes and trends in this data and it can help you to prioritise where to make improvements.
#2 Crazy Egg
Crazy Egg enables you to gain a greater insight into the behaviour of users on particular website pages than you can with Google Analytics.
Once integrated with your website, you can track what users are doing on the pages you want to investigate using the Crazy Egg heat mapping technology. The reports you can generate include options for seeing the intensity/number of clicks on different areas of the page (heat map view) as well as the percentage of users that viewed whole horizontal areas of the page (scroll map view). You can also distinguish clicks by the traffic source to the page (Confetti view).
You can also track behaviours of users browsing the site on different devices (desktop, mobile, tablet). This is great for understanding if your responsive or mobile site is actually optimised for a mobile device – are the page layouts and content appropriate for mobile browsing behaviour.
Like Google Analytics, this tools helps you to discover what is going on but it also starts to help you to identify why users are behaving in certain ways.
For example, you might notice that hardly any users are clicking your call to action (CTA) using the heat map view in the report. When you switch to the scroll map view it’s evident that users just aren’t scrolling far enough down the page to actually find the CTA.
You can now form a test hypothesis that if the call to action were placed in an area on the page that received a higher number of views, more people would see it and therefore click on it.
I mentioned earlier that we need to find out why users are behaving like they do on a website as part of our usability testing.
The best way to do this is to actually talk to them, to get the voice of the customer (VOC).
This doesn’t necessarily mean picking up the phone and calling them or talking to them while they are using the site.
A really easy and automated way to get feedback from your users is to ask them to answer a short questionnaire.
– What is the purpose of your visit to our website today?
– Were you able to complete your task today?
– If you were not able to complete your task today, why not?
You can also add the Net Promoter Score question to your free survey:
– How likely is it that you would recommend us to a friend?
By reading the explanatory links in this section you will see how these fundamental and powerful questions will lead to tremendous insights into how well you are meeting the needs of your website users and if you have any usability problems.
Based on this VOC feedback in combination with the information from the other tools listed here, you will be able to develop some great website improvements to test.
If you want to ask more questions in your survey, then you can pay for the service to do so.
#4 UserTesting.com (Peek)
Much like the iPerceptions survey tool above, UserTesting.com helps to gain VOC feedback and insights. It tells us the why.
This is a paid for service where participants are sent a link to your website or app and asked to explore it and perform certain tasks whilst talking about the actions they are taking.
This is brilliant honest feedback and is extremely helpful for gaining insights that generate testing hypotheses for UX optimisation and CRO.
You can gain some very interesting free insights by using the Peek user testing service.
If you do decide to pay for this service, the UserTesting.com team can help you to create and analyse your tests (for extra fees) or you can write and analyse your own questions and results.
I find it useful to create the questions/tasks for the participants based on Google Analytics and Crazy Egg behavioural observations, as well as the VOC feedback from the iPerceptions surveys.
I have mentioned above that it is important to test any improvements made to a website or app. The free Optimizely tool helps us to do that really easily.
Based on the behavioural observations and feedback from users we can develop optimisation hypotheses to test.
For example, when I was talking about Crazy Egg earlier I mentioned that based on behavioural observations we could infer that moving the position of a CTA to a more highly viewed area of the page would generate more clicks (more micro or macro outcomes and conversions).
We could test this out in Optimizely once the software has been integrated with the website we want to run a test on.
Once integrated, Optimizely recreates the web page you want to test within the Optimizely tool/platform.
You can then make changes to the test page (like moving the CTA to a different place) often by simply dragging and dropping content items (e.g. CTAs, images, text, media players, forms, etc.) to different locations. You can also easily change or remove text and images, and change the colour of the items.
The first time I did this it totally blew me away, it’s like magic!
The changes (hopefully improvements) you make should be based on the hypotheses you want to test. Therefore, as well as making the changes, you need to tell Optimizely what the actions you want to track are. For our previous example, it would the number of clicks on our CTA.
Then you can run an A/B split test where you send some of the website users (e.g. 50%) to the original version of the page and some of the website users (e.g. the other 50%) to the test page.
Over time, Optimizely tracks the user behaviour on the different versions of the page (you can simultaneously test several versions if you like) and tells you if the behaviour differs at a statistically significant level. It is gaining statistically significant results that gives your usability testing a level of credibility and robustness needed to then decide if the changes/improvements featured in the test page(s) should be implemented on the website.
If you have a lot of regular website visitors you will be able to reach statistically significant results faster, and potentially be able to run multiple tests simultaneously.
If you do not have many website visitors then you should only run one test per page at a time, otherwise, you could be waiting indefinitely for statistically significant results.
I would also recommend that you integrate Google Analytics, Crazy Egg, iPerceptions and UserTesting.com with your Optimizely tests in order to maximise the insights you get. If you don’t, then you will not really be sure as to why your tests succeed or fail.
In relation to my last point, be prepared to fail. Failure is an inevitable part of testing. But if you do your best to discover what users are doing and why they are doing it, you will have far more success.
I hope that you found this post useful, and enjoy the delights of behavioural research, usability testing and CRO.
About the author
Alex Hutson is one of Oxford College of Marketing’s lecturers, teaching at our centre in London and online through our extensive webinar programme for distance learners. Like many of our assignment tutors and lecturers, Alex is also a consultant through Website Ability.
If you would like to find out more about Alex you can connect with him on linkedIn at https://uk.linkedin.com/in/alexanderhutson