Last week we partnered with Contentsquare for a champagne breakfast event discussing ‘sexy experimentation’ where 5 optimisation experts spoke about conversion rate optimisation (CRO) strategies to boost digital businesses.
Our CEO Neil McKay was one of the speakers at the event held at The Humble Grape, Fleet Street, on 28 February. With 93 attendees, 5 entertaining speakers, and an upbeat Q&A session, the event sure got exciting.
The topic for the morning was rejuvenating the culture of experimentation among eCommerce businesses and focusing on improving user experience to drive your business goals. Here are 15 tips these CRO experts had for digital businesses:
Neil McKay, CEO of Endless Gain, spoke about the importance of emotions in a customer’s online journey. He discussed the significance of using biometrics and psychology alongside traditional research and analysing methods to get a full picture of what the customer is doing on the site, why that kind of behaviour occurs, and how the experience can be improved.
Here are the key takeaways from his presentation:
When a customer is on your website and scrolling through or moving from one page to another, they are actually aroused at different levels. It is important to understand these emotions as they have a direct impact on your sales.
Your customers’ emotions are either driving them towards you and that purchase, or driving them away from you and that purchase. By understanding them you are giving yourself the best possible opportunity of improving your customers’ digital buying experience, one that you will benefit from through sales.
Biometric research uses biological signals to understand how customers react to or feel about a website/app and how intense those emotions are. The methods used are: Facial Expression Analysis, Galvanic Skin Response (GSR), Electroencephalography (EEG), Pupil Dilation, and Eye Tracking.
Always start with data analytics, with traditional research methodologies such as on-site analytics, sessions recordings, historical test data, market research, scientific research, customer surveys, and functionality, accessibility, and usability tests.
Then, overlay it with biometrics to get advanced insights into why websites are not performing at the level they could or should.
Then, take the insights and use psychology in your design. There are 3 key psychological principles to use in conversion rate optimisation: Attention (how to capture the customer’s attention), Persuasion (how to persuade them to perform a particular action on your site), and Motivation (how to motivate them to complete the action).
Based on these principles, there are hundreds of psychological techniques to use that will help you give your customers “the best possible opportunity of an emotionally enhanced buying experience”, one that your business will benefit from.
Any CRO strategy you use will underperform if you do not pay attention to the functionality, accessibility, and usability of your website. By merely making sure that your site is functional and accessible to all kinds of users on all browsers and devices and that every part of the site is easily usable, you can recoup a lot of pounds in revenue that has been leaking from your site each year.
Watch Neil McKay talking at the event.
Bhavik Patel, Head of Conversion at Teletext Holidays, presented on how it would be if Elon Musk did CRO. His focus was on the First Principles Thinking, which is the deconstruction of something to its fundamental parts that you know are true and build from there.
CRO is not really the goal businesses should be looking at; it’s a lever and the actual goal is revenue and unit economics. Patel broke down revenue into its fundamental principles or components, with this flowchart:
Patel broke down the customer journey on Teletext Holidays into several acquisition and retention touchpoints. He then illustrated how it is important to know these customer touchpoints first, before you decide on what pages or experiences you want to run experiments.
James Fearne, Customer Success Lead at Contentsquare, spoke about how to decide and prioritise what tests to run. Here are the key takeaways from his presentation:
This does not mean that you don’t involve the company/client/internal teams when looking for experimentation ideas. Fearne stressed on not wasting time and effort in asking everybody for ideas to experiment, but instead to focus on data-driven hypotheses.
Everyone should get involved and be encouraged to voice their ideas in order to foster a culture of experimentation, but the hypotheses should not be based on opinion, gut feeling, or organisational hierarchy.
“There are no stupid, wrong, or bad ideas; some ideas just don’t survive prioritisation,” he said.
Experimentation doesn’t always have to involve weeks of running tests and waiting for statistical significance. If you want to try something different and have got the right data, go for it.
When experimenting without running a test, conduct early analysis of your campaigns and implement the changes backed by data.
Sometimes, Fearne says, your experiments are not user-focused; they’re just hypotheses you want to validate. Unless you have more than one idea and your hypotheses are rooted in data, your experiment is not really a test. True experimentation starts with coming up with multiple user-centric solutions.
Marcello Pasqualucci, Head of Optimisation Strategy at HSBC, spoke about why it’s important to always keep the user in mind and why getting too smug with experimentation is not good for growth.
There are two CRO models prevalent in the industry: a linear support model where one account manager and team deals with one product or client and a user-centric model where you look at products/clients in terms of their customers.
While there are pros and cons to each model, the user-centric model works better, because it gives global alignment to the brand, personalisation of content for users for better engagement, and constant learning and growth within the team.
Pasqualucci stressed on the need to gather and analyse/understand offline data (non-digital behavioural mapping) and temporal data such as how a customer feels about certain words or colours, socio-economic patterns and local insights that might affect user behaviour or purchase decisions.
While you should never get too comfortable and hesitate to conduct a new or different experiment, make sure the tests are based on science and data. Pasqualucci spoke about behavioural testing, end-to-end experimentation, applied neuroscience, and the use of social physics in conducting tests. Social physics uses mathematical tools/equations inspired by physics to understand user behaviour.
A web experiment is not always simply A/B testing. Look beyond that and see what you can do at each of the customer journey touchpoint to enhance their experience. If you give a customer the right motivation and speak to them based on their needs, they will react positively to your business and products.
Octavia Benham, Online Trading Manager at Dreams Beds, described her online and offline testing experiences with what she called the “Goldilocks’ guide to testing”. Her focus was on customer feedback, understanding that there are limitations to testing, and setting up a clear testing roadmap.
Getting direct feedback from customers is very valuable—it drives testing, inspires you to do things differently, and think like them. Benham said that it’s important to get customer feedback at every step of their journey, whether in-store or online, including delivery. This gives you a holistic view of customer satisfaction.
Not everything can be tested, and each test may have limitations—either from your end, from resources or traffic bandwidth, or from the user end. So, focus on conducting data-backed tests that have a higher chance of providing a positive outcome. While you should dare to be different and try something new, make sure it’s doable.
At the same time, there might be inconclusive tests where no significance difference is seen between control and variations. This should not be something to be disappointed about, but rather as an eye-opener, because that page might have other elements—content, CTA, text, image, background colour, or layout, which is not exciting the customer. So, learn from your experiments and apply the learnings to the next tests, she said.
Create a feasible hypothesis based on data before you even begin testing it, works better. When you have some positive data to base your testing ideas on, you will have a clear idea of what you want to test and when.
Instead of going for 15 large tests a year, go for 50 smaller tests, Benham said. This will allow you to test multiple parts and elements of your website instead of getting limited to a few pages. She also stressed on the need to create a testing roadmap with clear KPIs and goals by brainstorming ideas and building a plan of action for the next few months. This will make sure that you get buy-in from business stakeholders and get more done through the year.
Enjoyed reading this? Maybe you’d like watching the videos of the event as well.
Here at Endless Gain, we improve brands profitability by removing blockages in their customers digital buying journey. If you’d like to know more please do get in touch with us.