Cashier-less shops are not all they are cracked up to be as the emerging bloodbath in China and Amazon’s glacial roll-out clearly indicates.
Amazon was first to launch a cashier-less store in 2016 but the fact that it has not taken this concept any further is a clear indication that this experiment was unsuccessful.
However, that did not stop a gung-ho roll out of cashier-less stores during 2017 and 2018 in large Chinese cities.
In Guangzhou, i-Store rolled out a chain of un-manned convenience stores while JD.com announced plans in July 2018to offer 5,000 unmanned kiosks in office buildings and train stations.
Buy-Fresh Go also began a roll-out in Shenzhen in 2017 of stores with no cashiers.
Given, the prevalence of easy to use mobile payment systems in China, the expectations were that this would quickly take off but. as always, the devil is in the details.
It turns out that there are several difficulties with cashier-less stores which have made them both unpopular and unprofitable.
First, Vending machines: the simplest way to create a cashier-less store is to sell pre-packaged goods, processed food and drinks which have a long shelf life and come with a barcode already attached.
The problem with such stores is that they are little more than a range of vending machines which have been prevalent across Asia for decades.
Hence, they offer little more than what existing vending machines offer, making them uninteresting and not an improvement on that which is already widely available.
Second, Margin: in order to be sustainable in a large city, one of these stores needs to generate between 5,000 to 6,000 RMB per day to survive.
Long shelf life processed food which is easy to sell automatically has gross margins of around 25% while fast food and fresh groceries have gross margins of around 45%.
Consequently, the more one can shift the revenue balance towards fresh food and junk, the more sustainable the store becomes.
Unfortunately, these products have short shelf lives and often require more than just a vending machine to sell to customers.
Third, AI: Unfortunately, machines are too stupid to cook burgers or serve up fresh groceries meaning that it is very difficult to offer the high-margin products in a cashier-less store.
Ideally, there would need to be some kind of robot present that is able to deal with this as well as fix any problems that arise.
Given the state of AI today,we are pretty far from this meaning that to sell high margin groceries requires humans.
I suspect that these issues corroborate with what Amazon has found which is why its roll-out of this proposition has been very slow.
Consequently, with many of these stores being little more than vending machines, their usage has not taken off and the profitability has not been high enough to make them sustainable.
The result has been a bloodbath with many of these convenience stores closing and ambitious plans being quickly cancelled.
Unsurprisingly, the last man standing is Alibaba which did not get in on the craze, but instead adopted a slower approach to retail automation.
It is rolling out a supermarket chain called Freshippo which sells fresh groceries and has much greater automation but also there are still humans around to do the jobs that machines cannot.
This is a much more sensible approach and further underlines my view that Alibaba is both getting much better at understanding the limitations and uses of its data and AI
For everyone else, large investments have been lost most of whom have now retired to lick their wounds.
Alibaba is the only place I would want to have exposure to China retail both for the online and offline environments.