00:00:00
Dan Loney: Well, no doubt, if you've been
to your local grocery store or a
00:00:03
Target or a Walmart in recent
months, you have seen the rise
00:00:08
in the use of self checkout.
Companies have been mixing in
00:00:13
this technology because they
believe it to be cheaper but
00:00:17
also more effective for their
consumers. But it appears that
00:00:20
there are some issues out there
with self checkout, and in fact,
00:00:25
in some cases, retailers are
cutting back due to rising
00:00:29
levels of theft. So what does
this mean for the technology?
00:00:33
Pleasure to be joined by
Santiago Gallino, who is an
00:00:36
Associate Professor of
Operations, Information, and
00:00:39
Decisions here at the Wharton
School. He's also an Associate
00:00:41
Professor of Marketing. Santiago,
great to talk to you again. How
00:00:45
are
00:00:45
you, sir? Santiago Gallino: I'm doing well. Doing
well. Thank you very much for
00:00:48
having me. - Okay,
00:00:49
so this is an interesting
dynamic at play here, because of
00:00:54
what was believed to be just the
absolute ease at which
00:00:58
technology was going to help the
process of people getting in and
00:01:03
out of stores. But I guess, did
we have to maybe see something
00:01:08
like this coming when you have
to factor in the human component
00:01:13
with the shoppers as
well?
00:01:16
- Yeah, so I think this
is interesting, and as
00:01:20
you are hinting to, is probably
also expected to some degree. If
00:01:26
we think like, what are the
retailers thinking when they
00:01:29
decide to bring a technology
into the store, self checkout
00:01:33
being one of them, they try to
weigh the pros and cons. And I
00:01:38
think that in the
early days, it was very clear
00:01:42
that there was a potential labor
saving component that was to
00:01:49
compensate for the investment in
the technology, also allowing for
00:01:55
mediating the peak demand times
and making everyone happier with
00:01:59
lower wait times in the lines
and whatnot. Now, the unknowns of
00:02:05
that process are, on the one
hand, to what extent customers
00:02:09
are going to be frustrated that
now they need to do the part of
00:02:11
the job. And the other one,
which is what you mentioned
00:02:14
earlier, is like, to what extent
we're going to see intended or
00:02:19
unintended errors in the
checkout point. And that's what
00:02:23
we're seeing now, that once the
retailers are starting to
00:02:26
actually observe those unknowns,
they are kind of recalibrating
00:02:30
that decision.
00:02:32
- It's been mentioned that, I
guess with some retailers, they
00:02:36
are limiting the number of items
that people can purchase when
00:02:40
they go through self checkout.
And I guess the retailers see
00:02:43
this as a way to maybe mitigate
some of the potential problems.
00:02:47
- That's right. And I think that
this limitation on the number of
00:02:51
items speaks to this dual
problem. On the one side, the
00:02:56
complexity of the task itself,
like we are customers, not
00:03:01
expert cashiers. And although
the task is simple, you can make
00:03:05
mistakes, and these mistakes can
increase with more items. At the
00:03:10
same time, if you want to assign
one employee to monitor multiple
00:03:17
self checkout or you're going to
do it with a video camera, it's
00:03:21
more accurate and simple to do
if you limit the number of
00:03:25
items. So I think that the
simplification of the task, it's
00:03:29
helping the customer, and it's
helping the retailer to monitor
00:03:33
that step too.
00:03:35
- And this is not an instance
where companies are going to
00:03:38
rethink having the technology in
their mix. They're still going
00:03:42
to have it. It's just -- and this
falls right in line with
00:03:46
your work. It's an operational
process that they have to kind
00:03:50
of work through to see how they
can best run this now.
00:03:54
- Yeah, so I
think you're right, that for most
00:03:57
retailers, this is a question of
kind of recalibrating, not
00:04:01
taking the technology away. The
ones I've seen that have kind of
00:04:06
pushed into removing the
technology altogether, it's more
00:04:10
of an argument of missing the
connection with the customer,
00:04:15
missing that kind of opportunity
of interaction with the
00:04:19
customer, making the whole
experience too detached from an
00:04:23
environment where the customer
feel embraced. And so it is more
00:04:27
like, if you want like a more
of a strategic reason with the
00:04:31
companies that see these still
in the space of, okay, we can
00:04:37
make this work, is trying to
mitigate the problem by reducing
00:04:42
the hours of operation, reducing
the number of items, updating
00:04:47
the technology. I think over the
last few years, we've seen new
00:04:50
technologies like, now, sometimes
you have the possibility to
00:04:53
weight the product after you scan
it, and that itself as a
00:04:56
reinforcement mechanism, and
things like that.
00:04:59
- You mention about the
adding in of workers in and
00:05:04
around a self checkout area to
kind of help monitor it. You
00:05:08
also talked about the use of
cameras. I've seen it go as far
00:05:12
as my local Target here in the
Philadelphia area, to actually
00:05:14
have a security person just
outside the self checkout area
00:05:19
so that if there is a problem.
So the problem of shrink has
00:05:24
been one that retailers know has
been there, and now they see
00:05:27
this as a way to try and
mitigate shrink, at least in
00:05:30
this arena.
00:05:31
- Yes, no, absolutely. And shrink
is not a new problem in retail.
00:05:35
I mean, retailers for years has
been dealing with these problem
00:05:41
both from customers, their own
employees, and this is something,
00:05:45
that it's been embedded in
processes over the years. I
00:05:49
think that now, because of the
level that this issue has,
00:05:53
the level that we are
seeing of this issue is where
00:06:00
the retailers are making
additional investment once the
00:06:04
technology is in place. But to
your point, I mean, if you have
00:06:08
one employee monitoring two self
checkouts, then you start to
00:06:12
question, like, to what extent
this trade off of labor
00:06:17
for the equipment is actually
making sense, right?
00:06:21
- So then, do we look then to
advancing the technology even
00:06:25
further to be able to try and
help this process out? - To some
00:06:29
extent, yes, and that's why I
think there are now companies
00:06:32
that are offering some
integration with an AI video
00:06:36
monitoring tool that will kind
of politely suggest that you
00:06:41
made a mistake, that there is an
item that you forgot, that
00:06:44
these are the organic
tomato, not the regular tomato,
00:06:47
and so you're kind of making
those savings on the
00:06:50
checkout step. And I think
that that is something we're
00:06:53
going to see, in the same way we
saw the companies incorporating
00:07:00
in the weight of the product at
the end as an additional
00:07:04
checkpoint. I think we're going
to see more and more
00:07:07
refinements. But I mean, you
don't need to be too creative to
00:07:12
think ways around if your intent
is to cause the shrink, right?
00:07:17
- So now,
00:07:17
some people would say the
easiest way to get around this
00:07:19
is add more human beings to the
mix, but companies are
00:07:23
probably not going to do that.
00:07:25
- Well,
that's probably part of
00:07:27
the story. And I think that
that's why I mentioned that
00:07:30
companies that have not gone
down that path are retailers
00:07:34
that see the value of that
interaction, as something that
00:07:38
adds value to the customer, to
the shopping experience, and
00:07:41
that, I think, is why there are
good reasons to stick to your
00:07:46
strategy and not necessarily
jump on every opportunity of a
00:07:49
new technology coming to the
store. - But it
00:07:52
does seem like that the use of
this technology will remain, you
00:07:57
know, somewhat under control in
terms of the amount that they
00:08:00
use it. You will still have
checkout lines that have human
00:08:03
beings behind them in the cash
registers, because you don't
00:08:08
want to -- in many cases,
retailers, I guess, don't feel
00:08:10
secure enough to have totally,
you know, self checkout
00:08:15
operations in place.
00:08:17
- No, absolutely. And what
definitely is going
00:08:19
to happen, and it is
happening, is for those retailers
00:08:22
that have started to make
investment, remember, these are
00:08:25
companies that has, like, thousands
of stores. So to what extent
00:08:29
they continue to roll this out,
expand. I mean, now that they're
00:08:33
seeing the numbers that
are coming up for those
00:08:37
components that they didn't know
before making the investments
00:08:41
are now reasons why they will
probably think twice before
00:08:46
jumping in, assuming that shrink
is going to be of a level that
00:08:51
it turns out not to be, benefit
from the
00:08:55
customer side, from the employee
side, are not as high as they
00:08:59
thought. So I think you're right,
that we are going to definitely
00:09:03
continue to see the technology
in place, but I don't think
00:09:07
we're going to completely remove
the workers from the
00:09:12
checkout line anytime soon.
00:09:13
- Santiago, great to have you with
us today. Thanks very much.
00:09:16
- Thank you for having me. - Thank
you. Santiago Gallino, Associate
00:09:20
Professor of Operations,
Information and Decisions, and of
00:09:23
Marketing here at the Wharton
School.
