I apparently am different from many Britons. I totally love self-service checkout systems, particularly at grocery stores. Many Brits apparently have a different take — at least according to the BBC (The problem with self-service checkouts, Dec 9):
New research suggests 48% of Britons think self-service checkouts are a nightmare, neither quick nor convenient. Quite the opposite in fact, and their complaints are all too familiar.
Now the article is unclear about the other 52%. Are they indifferent or in love? Do they dislike the machines to an extent less than totally loathing? What is clear is that the author of the piece has an ax to grind and would like to sink that ax into the speaker squawking “unexpected item in the bagging area.”
Despite her flights of hyperbole, the author has a point. These machines are not perfect and take some learning to use well. She points to specific issues like not being able to use your own bag or that mysterious warning when the recently bagged item does not match what the machine was expecting. Apparently, Brits also have paranoid thoughts that have never crossed my mind:
Finally, after the palaver of paying, there’s the nervousness about leaving the shop. Did I scan it all correctly? Did I select the right type of bread roll from the menu? Will I feel the long arm of the store (manager) on my shoulder as I walk out the shop?
“I spend half my time worrying that security will arrest me for selecting the wrong price Blueberry muffin,” said shopper Sharon Adams when consulted in a survey on self-service tills conducted by Fatcheese.
So if so many customers hate them, it must be that the companies are saving a ton of money by using them, right? The firms claim there is more to the story:
Supermarkets say the move towards self-service checkouts is not all about cutting costs. They argue the tills can speed up your shopping trip, says Ahmed Zaman, from shopping website Fatcheese, which conducted the research. “But many shoppers have yet to be convinced that they really save time,” he says.
Not everyone agrees with this analysis, claiming that assertions of time savings are misleading because they are based on being able to walk right into service without waiting. Further some assert that consumers are fooling themselves:
“People perceive self-service checkouts to be quicker but that’s because they are actually doing the work,” says [Bjorn] Weber[, of retail analysts Planet Retail]. “In reality they take longer than someone serving you, but it’s annoying for the shopper to stand around waiting.”
This suggests that Mr. Weber is a fan of David Maister. Back in 1985, Masiter published The Psychology of Waiting Lines, which laid out some principles for managing waits in service settings. One of his main points is that “Occupied Time Feels Shorter Than Unoccupied Time.” So it is not clear that trading a wait for traditional service for a more engaging process of checking out is per se bad. Also, I am not sure that it is wrong to focus on settings in which customers can just walk into service. In the very first post of this blog, I argued that an advantage of self-service systems is that they can provide extra capacity at off-peak times when traditional staffing cannot be justified. To put this another way, when my local Jewel is totally packed on Saturday afternoon, they would be better off with two conventional checkout lanes in place of the four self-service lanes they have. However, late in the evening or early on Sunday morning. They can provide more capacity and a better chance of walking right into service than the Dominick’s two blocks away that does not have self-service checkout. I once overheard a manager at Jewel comment to another worker that he had not had to man a cash register since the self-service lanes had been added.
Let me add that the supermarkets in Chicago (OK in greater Evanston) don’t use self-service lanes correctly. Most stores don’t regulate who utilizes the self-service options. I can check myself out even if I have 100 items in my cart. Here’s a phrase I rarely use: They did this better in North Carolina. I usually save that phrase for pork barbecue but it applies here. My first experience with self-service checkout was at Harris Teeter stores in North Carolina. They reserved the self-service lanes for those with a small number of items. One can break down the checkout process into two steps, a fixed time to process payments etc and a per item scanning time. (We’ve done this. See this post.) The fixed part is going to be largely the same between standard and self-service checkout — particularly when it comes to credit card payments. Customers, however, are slower than pros at scanning cereal boxes and identifying Fuji apples. Hence, the store loses capacity when it moves customers to a self-service lane (assuming that a full-service lane would have been staffed). How big a hit they take is driven by the number of items in the basket. My local Jewel has four self-service checkout stations AND at least one if not two express lanes for those with small baskets open. That’s boneheaded. Those who are the best fit for self-service are being served by an expensive resource that is more efficient with larger baskets. As I said, they did this better in North Carolina.
Great BLOG and some of your other BLOGS. I truly appreciate that an express lane can create frustrations for “express customers” when behind a larger basket.
That said, what do you think of the hypothesis – especially since setup is the larger part of time for below say a 14 piece orders – that express lanes overall increase checkout time overall?
I believe that for an order of any size a customer will be processed faster by an experienced cashier than by doing it on his own in a self-service lane. I use the self-service lanes at my local stores a lot. (Assume I am asocial and prefer not to make eye contact.) I often have small orders with items with obvious bar codes (e.g., a box of cereal or a package of ground beef) so I can do the whole process pretty quickly and don’t have to fuss much with the bagging. Still I bet that a trained professional would be faster. (However, there is a confound on this. For the trained professional, we would likely just be counting the time to scan my items and process my payment. I still have to do the work of unloading my cart and we may need another person to help with bagging.)
Even though the pro is faster once you are in service, I do think that self-service lanes make customers better off. As I argued in the post, this is especially true at off-peak hours.
I think what Brett is asking, is “Do express lanes (manned 15 items or less) always make service worse than the same lane with no restrictions?”
Let me take another shot at this. Let’s take some numbers to make things a little more concrete. Suppose that there are 42 seconds worth of work per customer regardless of how many items there are. This is the “set up” time and accounts for greeting the customer, processing the payment etc.
Now suppose that a small basket shopper has 6 items and a big basket shopper has 26 items and that it takes 3 seconds per item to scan and bag each piece. That means that a clerk spends 42 + 3*6 = 60 seconds on each small basket shopper. A big basket customer takes 42 + 3*26 = 120 seconds. Thus if there is only one clerk, he can process 60 small basket customers per hour if that is the only type of customer he sees. On the other hand, he can process only 30 big baskets per hour.
At first glance, that looks like small baskets are good. However, if we measure output in terms of items scanned per hour, big baskets win 30*26 = 780 to 60*6 = 360. So a clerk who works only on big baskets is going to be having a lot more cash going through his till than one who is only working on small baskets.
Let’s suppose that at a given store one third of customers have small baskets. Then the average customer take (1/3)*60+(2/3)*120 = 100 seconds. A clerk could then handle 36 customers per hour. Suppose our store has 3 clerks. They then have the capacity to handle 3*36 = 108 customers per hour.
But suppose they have an express lane. They have to commit at least one clerk to that lane and, hey, they are assigning one third of their resources to one third of their customers. Suppose that they are seeing 99 customers per hour. When there is no express lane, the utilization of the clerks is 99/108 = 91.67%. However, when there is an express, the express lane clerk’s utilization is just 33/60 = 55%. In the regular lanes, we get 66/(2*30) = 110%! That ain’t going to work, the big basket lanes are going to be overwhelmed.
What’s happening here? Capacity is lumpy; I can’t fractionally split the lanes. On top of that, when I move a clerk to the express lane, his capacity jumps by 67% (from 60/36) but those left with big baskets see their capacity fall by nearly 17% (from 30/36). So in a nutshell, as this example shows, having an express lane can cost the system some capacity and it is likely to be “unfair” in the sense that the express clerk will be less heavily utilized than those left with the big baskets.
[...] aside from Fresh & Easy, they have not completely squeezed out human cashiers. As I have argued before, grocery stores still need to staff up for peak times. The real benefit from self-service checkout [...]
[...] into those who love self-checkout and those who hate them. (We already discussed it here: “Love or loathe: Self-service checkout” where my colleague confessed his love to [...]
[...] have previously stated that I love self-service checkouts. It may just be that I hate making eye contact (ask anyone who [...]