Unlocking AI and Loyalty Solutions
Want to learn more about AI and Generative AI? Want to understand how to get started or improve on what you are doing today—watch this video courtesy of Netail and Ascendant Loyalty Marketing. Dr. Mark Chrystal and David Slavick share practical insights that you need to know.
You Might Also Like
Hi, I'm Mark Crystal. I'm the CEO of NetTail. And it's my pleasure today to
have this session
with David Slavik talking all things AI and so I'll let David introduce himself
Hi, David Slavik here, co-founder and partner at Ascendant Loyalty. We're a
full-service CRM
Loyalty Consulting firm. And actually, I've known Mark, believe it or not, this
is 2024.
We've known each other for 20 years. Because Mark and I worked together at
American Eagle
Outfitters when I headed up Loyalty and he was like all things merchandising
and planning and
driving that business forward. So Mark, it's really great to visit with you
today and
just for a point of disclosure, the two companies are strategic partners. And
there's a reason for
it and that's what we're going to get into it today. To be honest, I was at
ShopTalk a couple
weeks ago in Vegas. And the number one topic that people came forward with that
they want
ShopTalk, and there was like 12,000 people there at the Mandalay, is tell us
about generative AI,
about AI, show us practical examples of how we can employ it to drive our
business performance.
So when I wind up doing these type of broadcasts, I always delve right into the
practical. So that's
what we're going to do right now. Mark, what is AI? What is machine learning
and how can it be
applied to drive client business? Yeah, well, why don't I just start with
answering what AI is,
or artificial intelligence, and then we start high and then we'll dive into the
practical
application. So artificial intelligence refers to the simulation of human
intelligence in machines.
And basically those machines being programmed to think and learn. It's a hot
topic, as you just
mentioned, really because it has the potential to revolutionize a wide range of
industries,
from healthcare to finance to retail, by improving efficiency, automating tasks
solving complex problems that are difficult for humans to tackle. And
additionally, advancements
in AI technology, especially since November of 22 with open AI's release of
chat GPT.
It's called the the public's imagination and kind of sparks renewed interest in
AI. The AI
term has been around for a while. But I think now people are just realizing not
only publicly,
but I think in businesses, oh, there's a real practical application of this.
And if I'm not
doing it, I'm going to be left behind. And so everybody's really focused on it,
even though
not everybody knows how to apply it, what it is, how to use it. And it's the
purpose of
us digging in a little bit deeper. And being gentle about it too, right? I mean
, you've got to hold
a client's hand through the process. And like, how do you get started as well?
I mean, like,
where do you think foundationally people need to just think through? I mean, I
have opinions in
that regard, having been on that side of the desk client side, I know where I
would start.
But where do you see the best way to get started? What are the questions that
you ask clients to
really know whether they're prepared for it and how you can best need them?
Yeah, that's a great question. So I have a lot of retailers come to me and say,
help me build an AI strategy and watch it by AI strategy B. And my advice to
them is actually
AI is just a tool to accomplish your business objectives. And so let's start
with your business
strategy. What are you trying to achieve? And then let's find some applications
where AI makes sense.
And it might be that they don't make sense based on your objectives and we'll
table the AI
till next year. But very often what we find is that if you talk to management
in retail companies
and other businesses, very often they have a list of pet peeves or issues that
they'd like to see
solved and certainly objectives that they're trying to achieve that are really
hard. And so a great
way to start is come up with three, four, five opportunities or objectives that
you're going
after as a business and then sit down with someone who's an AI practitioner and
expert and figure out
if AI is and can be applied into those business objectives. And then if so, how
you then go about
that? That a lot of people start with like, well, do I have the right data? And
at the end of the day,
this all starts with actually, do I have the right business problem? Do I
understand what I'm
trying to achieve? And then, and then figure out the AI aspect of it to get
started. So how do I
apply it? Right. So how to apply it? Because a lot of times we'll advise
clients and we'll do the
simple four box model. That's just here's where the revenue lift can come from
and the horizontal
axis. Here's the demand on resources in the vertical axis. When you're looking
at it,
they shouldn't say to you, build me a strategic rationale for why I should use
AI. They already
are using analytics. They already have data, but you hit the nail on the head.
It's a tool,
it's a solution, and it's kind of a step up in terms of what their capabilities
actually are.
Correct? Maybe. Again, it's a tool. So it's about like the right tool for the
right job, as my father
in law always tells me. So don't hit the nail with the sledgehammer either,
right? Yeah, exactly.
And there's just no reason to do it. There's no reason to do AI for AI sake.
Like, what are you
trying to achieve? And then if you think there's an application of AI, bring in
an AI practitioner
who really understands retail, who understands your business, can understand
your business objective
and think through with you, does it make sense to apply AI here?
So after you go through that initial consultative conversation,
how do you then get started? So do you always recommend to the client that they
should have
an advocate for the initiative? And where does that advocate typically come
from? What background
should that advocate have? It's also always best to have that advocate be in
the business.
And it can be an IT, but I found for the most part, the business objectives are
usually business
focused. And it's much better to have a business leader who is the advocate to
say, okay, let's sit
down and figure out if there's an application of AI at the end of the day,
those business teams are
the ones that are going to need to use whatever solution is developed. And so
they ideally need
to be involved from day one, like defining the problem, and then working with
whoever's going to
help develop the problem or find the solution to the problem and just be
working hands at hand in
hand. And I find that if you do that, the solution that you develop ultimately
usually ends up being
much more successful and more widely used than if it's coming in via IT and
they're kind of pushing
it on the business. Right. And how do you caution a CEO? So a CEO is looking at
CNBC, and there's
people being interviewed, and they're saying, you know, we're using generative
AI to
influence certain areas of our business by understanding, you know, the
language that's going on or to
build a more intuitive chatbot as an example, as opposed to looking at the data
, seeing the
performance of a business like a grocery business, which I saw you demonstrate,
you know, which is a
great one because of how many skews they have, the variability of geography,
the demand in the
grocery areas every day. There's multiple trips from customers, they're
identifying what 98%
of all customers that are shopping by matching their phone number to their
actual basket. And then
that practical application. So when you think about your solution called profit
mind, and you
demonstrate to a client how that that license solution can influence their
business decisioning
every day, how do you walk them through that? Do you want to show us something?
Yeah, absolutely. Happy to do that. And so just to kind of precurs, as a profit
mind is an AI
business analyst. So today, what's happening across the retail space, and it's
actually an
amazing stats, a retailer spend $198 billion a year globally on analytics
solutions.
Only to end up doing all of their key decision making in Excel spreadsheets,
which is incredibly, you know, error prone and time consuming. And so what we
've built is a
business analyst that's an AI that actually gets the teams ideally out of that
Excel
drudgery. It goes through it looks at the marketplace, it looks at the internal
data that
they're having to pull out of all these systems and put into Excel, it does all
that work effectively,
and then it goes through and figures out what are the opportunities to drive
incremental revenue
and incremental profits, how can I free up some working capital to fund growth,
and really get
those business teams moving much faster. So we design profit mind to basically
report like an
analyst reports, and so it produces an executive summary. It shows you what it
's found. So here,
it's found $980,000 in incremental revenue this week. It's found $1.3 million
in incremental profit,
and it's also found $46,000 in working capitals. It's basically cash that's
tied up in inventory
that can be freed up to fund growth. It's found opportunities like 237
opportunities across the
business. It's constantly scanning the web looking for new competitive products
that are released,
and it's found 72 new ones that it's asking you to take a look at. In this
example, it's analyzing
50,000 skews from a retailer. It's found a competitor for 92% of them. We even
analyze why we can't find
competition. Sometimes they're totally unique. In this case, it's tracking 24
competitors and
about 150,000 products in real time, competitively. So what it's doing is it's
taking in the business
objectives that you have, and you tell it in language what you're trying to
achieve, just like you
were with a human analyst. It then goes and collects all this external market
data, tracking
products, skew by skew. It then goes through and looks at your internal date
and starts making
recommendations. Here in this executive summary on the key insights, it gives
you a verbal description
of what it's finding. It gives you a tabular view of what it's found. It's
found a bunch of
price increase opportunities, and it's quantified what those are worth. It's
found some areas where
investment could be increased. Those are areas where you might be out of stock
or under stock,
and it's quantifying how much incremental revenue and profit you could drive if
you fund those areas.
It's also identified areas where you should reduce investments on over-stocked,
and it even goes to
the next step of not only am I over-stocked, but how am I going to get out of
that position? Can I
affect my on-order, or can I run a promotion? In this case, it's identified a
promotion to run,
and how much capital will be freed up after you run the promotion, and whether
your profitability
will be impacted while you run the promotion. Then it also shows you price
decrease opportunities,
so areas where it's recommending you change the price. If we go and look at
some of these opportunities,
so it will show you in priority order the actions you should take. So as a
retailer coming in on a
Monday morning, it's gone through all of that Excel work, basically pulling all
the data together,
and then quantifying the opportunities and showing you in priority order the
actions you can take
to drive the business. In this case, these are pricing opportunities that it's
found. The first
one's worth $286,000 for the business. You can click on any one of these
opportunities,
and actually in those opportunities it tells you the insights of here. In this
case,
it's actually found that in this category, you're significantly cheaper than
your competition.
You're actually about 16% cheaper than the competition. It's run a scenario and
figured out that if you
increase your prices by about 14.5%, that you'll only lose about 4% in unit
demand. So it's estimated
then that you can generate about 200,000 incremental revenue, and 286,000
incremental profit.
It gives you the action it wants you to take. It gives you deeper reasoning for
every single
insight it finds. It shows you a distribution of the price changes that it
would like you to take.
It even shows you item by item, the current price, the suggested price, the
impact that if you follow
its recommendation, it will take. You can even come in and have a dialogue with
the analyst.
So this is where you can come in and actually ask questions of,
do I want to run a different scenario or do I not like this scenario? And so
the example I always
like to provide is I can say, I don't like this price increase recommendation.
Or the increase is too high, judgementally.
Let's try. I don't know, David, do you want to suggest an alternate with a 10%?
It's too reasonable with our shoppers.
I've even misspelled and hopefully can read my misspelling. So what profit mind
is doing here is
it's understood we're asking it to rerun the pricing scenarios and so it's re
running, it's
recalculating every single skew. And in a few seconds, here we go, it came back
. It's recalculated
every item. It's recalculated now with that 10% cap. Here's my sales and margin
. It still doesn't
want to change every item by the full 10%. It's actually now said there's 57
skews instead of 58
that would be profitable with this change. And those are the big traffic
drivers and you could
get cross shopping out of it or something else as a result. Yeah. Yeah, exactly
. And it's gone
through and recalculated every single item. We've even told it that we want it
to have a 99 cent
ending in our rules. And so you can see it's converting that as well. And so
you basically
ask profit mind anything. It works just like an analyst working for you and
does lots more
besides in terms of understanding the business, giving you insights.
Really cool part is you can do all these what if scenarios. You're responding
back and forth in
real time with the tool. It's providing strategic written recommendations,
which is one of the
biggest gaps I think in the whole analytical spaces. You can show me all the
dashboards in the world
and give me all these snapshots. But then someone has to come up with the
rationale as to why.
And the fact is you've got that built in as well. How many months does it take
to set up something
like this with that many skews and also set up the scraping for the competitive
and as well?
Yeah. So we design this to be really fast. So usually up and running in four
weeks on the
competitive insights. And then within eight weeks on everything you just saw in
terms of the pricing
and the price optimization models and all the explanations and insights. So
normally within
two months we're up and running and providing strong value into the business.
It's amazing. It's just amazing. And now that you have the solution in place,
which just demonstrates
the power of the business as a partner. Are you doing proof of concepts? Or you
go from proof of
concept pretty quickly to an ongoing... Yeah, we operate either way. So we have
lots of retailers
that say, look, we just need help and let's just get going and let's skip the P
OC and let's just
jump right in. And we obviously love that. But also we'll do proof of concepts.
And
actually we found with all of our POCs so far that every retailer has seen
value and wanted to move
into a full subscription with the platform, which is really exciting. Yeah. So
our plan and
ascendant is to advocate this solution across all the different retail clients
that we service
and really bring those demos to them so that they could see the power of what
you bring.
So kind of like in closing, what do you see for the future of profit mind and
what do you see for
the future of AI besides just people getting more and more comfortable with it?
Is there something
that's happening next that we should be aware of? Yeah. And what's next is
going to be...
So what we've seen recently is large language models. So everything that's kind
of language-based
has been revolutionized with AI. We're going to see that move into the vision
space.
Is the next wave. You're starting to see that with technologies like OpenAI
Sora,
which is the video technology that OpenAI recently released. But there's going
to be lots of other
applications of vision-based technology, which for example could be cameras in
the stores as an
example looking at the shelves and identifying stockouts and helping design
teams look at products
and redesigning the products using a vision technology. Any application of
vision is coming
next. With profit mind, we're continuing to expand. We see it as an all-encomp
assing
analyst for retailers. So we've started with the core profitability decisions
around pricing and
assortment and an inventory. But we're ultimately going to extend it into
perhaps supply chain,
into customer analytics and other areas that AI is getting applied in the
retail space. We
think we have an analyst here that can come in and actually help supercharge
the teams across the
board. And so that's what's next for us. We are using some vision technology
where we're when we
do our competitive matching, we're actually looking at the retailers' images
and their products
visually and comparing them across the competition. So we can find fashion
products and competitive
products across fashion that doesn't have an exact duplicate out in the web. We
can find
very similar products and we look at it, basically our vision models look at it
the way a human would
look at it and the way they would shop. And so we've already got some of that
in and we'll continue to
extend that capability. So just as an FYI to conclude, Mark and I are drafting
up a white paper.
I'm calling it AI and generative AI and machine learning for dummies. Mark's
making sure that
I'm not a dummy when I'm writing this thing up and being a great editor, even
though I'm a pretty
good editor, but he's just beating me up as a result. We're on the fourth
revision because it's
so much to learn and apply, which is great. And so I'll be putting that out on
our website.
Mark will be sharing it out as well on his LinkedIn and there'll be links to
both of our companies
as a result. And it's really our pleasure to share this with you all on YouTube
and put it out there
for everyone to know and learn. But the profit-mind solution is just really a
game-changer in the
space and I encourage you to reach out to Mark. I don't know, Mark, if you want
to
provide your email address simple enough at the end of this and contact
information and people
can reach out to you. Yeah, we'll do that on profitmind.com and David really
appreciate the
partnership and the endorsement. It's our pleasure. Here's to another 20 years,
my friend.
Yes, absolutely. Totally agree.