At Econsultancy Live, Gregoire Baret – Senior Director of Omnichannel Experience at Aldo Group – shared how the footwear retailer has approached experience design in 2020, and what testing has revealed in a year that has been “a reminder of our omnichannel essence”.
Baret describes experience design as all about “designing services to enhance the shopping experience. What we mean by that is evolving the way you connect with consumers through ecommerce – whether it’s redesign, new features you want to add, better ways to connect ecom and stores,” he explains. “It’s also about evolving the store experience and enhancing the way you interact with your consumers in-store.”
Implementing ‘design Darwinism’
So, how can a brand like Aldo find the right opportunities, without getting bogged down in testing too many ideas at once? Baret advocates what he calls “design Darwinism”.
“The idea is really to find ways to establish – through tools and process – what are the ideas, services, or future, that are going to survive through ‘selection’ and survive to fit with specific performance criteria. We are trying to compare and score everything we do and we use a very simple process for that.”
Aldo uses the ‘RICE priority score’ – reach x impact x confidence / effort – where reach is how many people you are possibly impacting; impact is about how much it is moving conversion; confidence is to what extent you know it is going to work; and effort is the technical and operational challenge and how complex it is going to be.
This creates a simple calculation to prioritise design changes.
Reducing cognitive load and doubt
Aldo uses a variety of tools, ranging from AB and multivariate testing to consumer feedback, in order to uncover insights about the retail experience and design.
“We can’t rely on our intuition about what is a good or wrong design, we used to say to our designers ‘don’t fall in love with your design’ because we are going to expose the design in the context of reality, and to tweak it based on what people actually do.”
Baret cites an example of one particular A/B test, which involved comparing a standard ecommerce page (containing features including reviews and a sizing tools) with a ‘naked product page’.
“The main idea of an ecommerce page is to reduce any doubts. So, we were expecting the page [with all the aforementioned features] to be the best performer. Surprisingly, we found the [naked page] to be the best performing one.”
“The more we were to add some of these new tools, the more we were adding cognitive load and friction and the more, in fact, we were adding doubt.”
“The very moment where I am enticing you to check this size tool, this is not only giving you guidance, but it is questioning you – ‘is it the right size?’ or ‘is it not the right size?’” explains Beret. “When looking at the reviews, they were also leading consumers to question the reality of those reviews and what is the truth. Strangely, all of these tools that are known to be essential in ecommerce ended up not performing that well.”
“It doesn’t mean that we don’t need these tools, but the way they were implemented was wrong, and that we need to find a way to simplify and elevate some of these cognitive loads to make sure that we don’t infuse some doubt during your experience.”
Paying attention to what consumers are doing (not saying)
Baret admits that “..the better insights we do have are coming from the actual reality of implementation.” In other words, don’t listen to users, but pay attention to what they do.
One example of this credo in practice is Aldo’s testing of interactive screens in store to allow customers to discover products. Despite what consumers said to the brand when testing in the labs – e.g. “we like this tool, it is super useful” etc – the performance was not good in store.
Baret comments,“We kept on trying and thought ‘maybe what we need to do is have the tool introduced by the associate in store, and to get the consumer to interact with the tool for some very specific benefits.”
“So, each time you come in store, if you need a product we will be able to quickly check if it is available, and if so, we can quickly a request a runner in the back store to bring you the product. If the product is not there, we can showcase some recommendations.”
The idea behind the tool is that it reduces the frustration of waiting.
In the end, after iteration, the team found that comparing the initial fixed touch screen with a new mobile device, the latter was found to be far more consistent in performance and went on to grow in usage over time.
On the back of the test, Aldo rolled out mobile devices in all stores across North America. “This totally changed the way we do business, in our ability to be faster with consumers but also to capture data about the level of consumer interest in products – even if they don’t purchase,” says Baret.
“I have a new signal that is informing performance measurement that I can use for new experiments, that are also informing future strategies for inventory and location.”
Explore, iterate, test and learn
“This was one of the key learnings that we have generated through these type of pilots, in how we should explore, iterate, and test and learn to improve the service model in store,” says Baret.
He explains how it is crucial to consider biases, such as ‘lip service’, meaning that people don’t always do what they say. “They might tell you that they love a feature and that they would interact with it, but you need to test it to understand what is going on in reality.”
Other biases can stem from the geography of the store, the newness effect (meaning expected levels of excitement to begin with before a drop off), form factors, timing, or immediacy vs. delay (increase in sales but detrimental to equity).
