WEBVTT

00:00:00.041 --> 00:00:02.168
- The AI technology
definitely brings a lot

00:00:02.168 --> 00:00:04.796
of new opportunity to retail,

00:00:04.796 --> 00:00:08.133
but I also think it
actually amplified the need

00:00:08.133 --> 00:00:10.593
to have good merchants,
good product owners

00:00:10.593 --> 00:00:14.139
who can actually describe
what they expect from AI.

00:00:14.139 --> 00:00:15.390
- Crystal, Nick, welcome.

00:00:15.390 --> 00:00:17.851
Crystal, where is AI making
the biggest difference

00:00:17.851 --> 00:00:20.895
inside DFI Retail Group right now?

00:00:20.895 --> 00:00:23.356
- We're indeed embarking a journey

00:00:23.356 --> 00:00:26.568
to drive transformation with AI.

00:00:26.568 --> 00:00:29.112
So, if I can just talk about two aspects.

00:00:29.112 --> 00:00:32.240
One is we want our team
members to be AI savvy.

00:00:32.240 --> 00:00:33.908
We're making sure that they have a tools

00:00:33.908 --> 00:00:35.994
to equip them on a day-to-day basis,

00:00:35.994 --> 00:00:37.370
but at the same time, we are looking

00:00:37.370 --> 00:00:40.832
for some really high-value use case

00:00:40.832 --> 00:00:42.709
to make sure we are injecting AI

00:00:42.709 --> 00:00:46.171
into our merchandising process,
our supply chain process

00:00:46.171 --> 00:00:49.591
to make sure the right decisions
are made for the customers.

00:00:49.591 --> 00:00:52.719
- Nick, when you look
across the retail sector,

00:00:52.719 --> 00:00:55.597
where do you see the
biggest opportunities?

00:00:55.597 --> 00:00:57.223
- Yeah, I think the next frontier

00:00:57.223 --> 00:00:59.184
is really in the commercial functions,

00:00:59.184 --> 00:01:01.603
so take things like category
management and assortment

00:01:01.603 --> 00:01:03.521
where there's just so
much untapped potential

00:01:03.521 --> 00:01:05.899
with some of the things that
Crystal was speaking about.

00:01:05.899 --> 00:01:07.776
Take, for example, assortment,

00:01:07.776 --> 00:01:10.236
which really has a huge
potential to localize

00:01:10.236 --> 00:01:11.988
what we're offering customers,

00:01:11.988 --> 00:01:14.657
but historically, we've had
a real explainability gap.

00:01:14.657 --> 00:01:16.701
Being able to have the category managers

00:01:16.701 --> 00:01:19.496
understand what the science is saying

00:01:19.496 --> 00:01:21.414
and get that therefore
implemented in stores,

00:01:21.414 --> 00:01:23.083
and I think there's a huge opportunity

00:01:23.083 --> 00:01:25.043
to bring that together finally.

00:01:25.043 --> 00:01:26.961
- How do the improvements
that are taking place

00:01:26.961 --> 00:01:28.797
behind the scenes translate

00:01:28.797 --> 00:01:31.508
into a better experience for customers?

00:01:31.508 --> 00:01:35.178
- I think this whole AI strategy is aimed

00:01:35.178 --> 00:01:38.098
to help us build better
connection with our customer,

00:01:38.098 --> 00:01:41.059
so to do that, we need to make
sure we have the right data,

00:01:41.059 --> 00:01:43.061
the right people, and the right process

00:01:43.061 --> 00:01:45.897
to help us do better category planning,

00:01:45.897 --> 00:01:48.066
better promotion planning,

00:01:48.066 --> 00:01:50.235
space planning to be more personalized

00:01:50.235 --> 00:01:52.695
in the offer that we
provide to our customer.

00:01:52.695 --> 00:01:54.989
But all this comes with a degree of change

00:01:54.989 --> 00:01:57.784
on how human will operate with the system,

00:01:57.784 --> 00:02:00.954
so have to rethink about
the job description,

00:02:00.954 --> 00:02:03.164
the skills that we need
to operate with AI,

00:02:03.164 --> 00:02:08.336
and be very agile in the way
to learn and adopt as well.

00:02:08.336 --> 00:02:09.420
- You've said it so well.

00:02:09.420 --> 00:02:10.797
For me, it's really about being able

00:02:10.797 --> 00:02:13.341
to listen to the customer
through their actions

00:02:13.341 --> 00:02:15.135
and the intents they're showing us,

00:02:15.135 --> 00:02:17.303
and then being able to make
sure that we act on that

00:02:17.303 --> 00:02:20.265
in store or online so that
they feel that we're listening

00:02:20.265 --> 00:02:22.809
and carrying the products
that matter most to them.

00:02:22.809 --> 00:02:25.478
- Yeah, and I think
there's a long tail of work

00:02:25.478 --> 00:02:28.439
that human being never
used to be able to do,

00:02:28.439 --> 00:02:31.985
like doing a huge amount of data analysis

00:02:31.985 --> 00:02:33.903
to really understand the customer voice,

00:02:33.903 --> 00:02:35.780
the insight, the business performance.

00:02:35.780 --> 00:02:38.950
All these now with AI,
we have the opportunity

00:02:38.950 --> 00:02:41.286
to look at things in a more holistic way,

00:02:41.286 --> 00:02:43.246
in a much more scalable way as well.

00:02:43.246 --> 00:02:45.623
So, these are the things
that I think as human

00:02:45.623 --> 00:02:48.835
or very experienced
merchant, category manager

00:02:48.835 --> 00:02:51.963
will have to learn to do things
that they're really good at,

00:02:51.963 --> 00:02:54.799
which is making the right
decisions for the customers

00:02:54.799 --> 00:02:57.677
rather than trying to
analyze all the information

00:02:57.677 --> 00:03:00.138
that they have before they
come up with decision.

00:03:00.138 --> 00:03:01.723
- Nick, how do you make sure as a business

00:03:01.723 --> 00:03:04.309
that you don't drown in
all of this customer data?

00:03:04.309 --> 00:03:05.643
There is so much of it.

00:03:05.643 --> 00:03:06.895
How do you pick out what's relevant?

00:03:06.895 --> 00:03:09.147
- Yeah, well, the most
important work right now

00:03:09.147 --> 00:03:10.982
is to build the knowledge graph,

00:03:10.982 --> 00:03:12.942
you know, really codifying the information

00:03:12.942 --> 00:03:14.861
that's in retailers' heads

00:03:14.861 --> 00:03:16.738
and getting it into systems in a way

00:03:16.738 --> 00:03:18.323
that they can then help interpret

00:03:18.323 --> 00:03:20.074
and create some consistency

00:03:20.074 --> 00:03:21.534
in the way that that retailer thinks

00:03:21.534 --> 00:03:23.244
about their business and their format

00:03:23.244 --> 00:03:25.413
and the customers that
they're trying to serve.

00:03:25.413 --> 00:03:27.832
And so, it's about
codifying that knowledge

00:03:27.832 --> 00:03:31.252
so that when the data science
shows what the answers are,

00:03:31.252 --> 00:03:33.004
it can be connected to the intuition

00:03:33.004 --> 00:03:35.423
that the merchants really
have in their heads

00:03:35.423 --> 00:03:38.509
and be able to bridge
that explainability gap.

00:03:38.509 --> 00:03:39.344
- No, I agree.

00:03:39.344 --> 00:03:41.638
I think the AI technology
definitely brings a lot

00:03:41.638 --> 00:03:44.224
of new opportunity to retail,

00:03:44.224 --> 00:03:47.518
but I also think it
actually amplified the need

00:03:47.518 --> 00:03:50.021
to have good merchant, good product owner

00:03:50.021 --> 00:03:53.608
who can actually describe
what they expect from AI.

00:03:53.608 --> 00:03:55.526
So, that institutional knowledge

00:03:55.526 --> 00:03:58.154
becomes so much more important for us

00:03:58.154 --> 00:03:59.989
to make sure we're asking the AI

00:03:59.989 --> 00:04:02.325
to do the right thing for our customers.

00:04:02.325 --> 00:04:04.661
- Crystal, Nick, thank
you so much for your time.

00:04:04.661 --> 00:04:05.620
- Thank you very much.

00:04:05.620 --> 00:04:06.621
- Thank you.

00:04:06.621 --> 00:04:09.207
(gentle music)