This is the story of how a UK conversion optimisation agency agency, AWA digital, used an Ethnio web intercept, surveys, and experimental data to help a treasured tea shop in the brand increase web conversation rate. Their CEO, Dan Croxen-John has quite the story for you.
Bettys is well-known in the north of England for running the most fabulous old-fashioned tea shops, literally. Their customer service is out of this world, and so is the attention to detail. If you visit York, which is a really pretty city – the second-most visited city in England after London – you will see queues of people waiting to go into Bettys.
There’s always a queue outside. They have three or four other tea shops in different parts of the north of England. People queue there as well. It’s a real experience. York is also popular for stag and hen parties. Hen nights always start at Bettys. It’s a tradition.
We worked on optimizing Bettys’ website for three years. This insight came fairly early on. We used Ethnio to recruit research participants, and had recruited a visitor to the Bettys website. One of our team members, Dave, was sharing a screen with her and she was talking about the how she used the website, Bettys.
In the background, her boyfriend comes in, and Dave can hear this conversation. The boyfriend says something like “What are you up to?” and she says, “I’m doing some usability testing for Bettys.” He said, “Okay,” and then he said, “I didn’t know they sold online.”
It occurred to Dave perhaps other people visiting the website might not know that Bettys sells cakes and confectionary online despite the fact the websites display prices, has add-to-basket buttons, a perpetual cart and a free delivery threshold offer.
Dave found a way of using Ethnio to run a survey and we got it going that afternoon. Quickly we got over 400 responses to the survey and he found out that 25% of those people who answered the survey didn’t know that Bettys sold online. So, this triggered a thought. He said, “Maybe we should test adding some information just on the homepage to tell visitors you can buy online from Bettys." So, we added these nine words:
“Shop online now - from our bakery to your door.”
That’s all we added. We then split-tested this against the homepage that didn’t include that text. We got an 18% increase in sales – site-wide. In fact, for visitors who saw the homepage, it was a 34% increase, but the way that we calculate our results is about site-wide improvements, not just on the page where the improvement is.
When Dave went in to present the results to the e-commerce team, they just said, “That’s rubbish, absolute rubbish. Everybody knows that we sell online.” He said, “Well, when we surveyed them, 25% didn’t, and then when we tested this ‘Shop online now, from our bakery to your door,’ 18% more sales were observed during the test.”
I think there are a number of points about this results:
- Understand the limitation of Analytics. Google Analytics is great, web analytics data can tell you a lot but it would never tell you that a quarter of your visitors don’t know you sell online
- Each business and their customers are unique. You can’t run this test on another e-commerce website and get the same results.
- Listen. It's about being humble about what you don’t know and being open to new insights. Often these flashes of insight come from talking to website users and really listening to them – and their boyfriends…
Without using Ethnio to help us recruit real visitors of Bettys website, this insight, and ultimately the sales increase Bettys saw would not have been possible.