Every AI Supply Chain Strategy Ran a Pilot. Under 1% Got Enterprise Impact.

Every supply chain team has run an AI pilot by now.

Very few of them have an AI strategy

There's a difference between those two things,

and it costs some real money

So over the next three episodes, we're gonna fix that with

help from Professor Brett Duarte from Arizona State University's W. P. Carey

School of Business, ranked number two in the country for supply chain

Today, procurement, supplier risk, contract intelligence,

and what happens when an AI agent joins your sourcing decisions

Let's get into it

Welcome to the Data Blades Podcast, part of the Retail Razor Podcast

Network, where we cut through the noise and get to the data-driven insights

that move your business forward.

I'm Ricardo Belmar

I'm Casey Golden.

Welcome to season two, episode 13

Casey, we're doing something a little different

over the next three episodes

We are.

We've got one guest, one theme, three conversations, and our

theme is AI strategy for supply chain, one of my favorites

And here's why we're spending three episodes on this.

Most supply chain teams have likely run AI pilots by now, and plenty of them work.

The harder part is the layer above the pilots, the strategy that decides

where AI belongs, who owns it, and what it adds, what it's actually worth.

What's the transformational outcome you're trying to achieve?

So we split it up the way the supply chain operates.

This episode is procurement.

Next one is operations, and the third, logistics.

Our guide on this magical supply chain journey for

all three is Professor Brett Duarte.

Brett is a Clinical Associate Professor in the NASPO Department of Supply Chain

Management at the W. P. Carey School of Business at Arizona State University.

A little context on that because who doesn't

appreciate a little extra context when you're getting fresh insights?

Arizona State is a leading public research university known for

innovation, access, and impact.

The W. P. Carey School of Business is the largest business school in the United

States with more than 25,000 students

Its supply chain management program is ranked number two in

the country by US News and World Report.

Brett also serves as senior faculty co-director of the school's Master of

Science in Business Analytics program, and he's one of the faculty teaching

the school's AI-enabled Supply Chain Strategy Executive Program this October

5th to 7th in the main campus in Tempe

Which is where the series came from.

Three days, three functional areas, and we get to preview all of it just

for you, our listeners and viewers.

And today we start where the money starts, procurement.

Supplier evaluation, supplier risk, contract management, and what changes

when agentic AI starts participating in decisions people used to make

in spreadsheets and meetings.

But before we jump in, let me tell you about the Retail Razor Podcast

Network sponsor, RetailClub.

There is still time to join us at the RetailClub AI Festival.

From the founders of Shop Talk and Grocery Shop, RetailClub AI Festival

brings together 2,000 industry leaders from September 22nd to 24th

for three days fully outdoors in Huntington Beach, focused entirely

on how AI is transforming retail.

Hear from 150 leaders shaping how AI is being used across the industry,

explore the technologies changing how retail works, and connect with

senior leaders from across retail, brands, technology, and investment.

Late registration is now available at retailclub.com/retail-razor-podcast.

We'll see you in Huntington Beach!

Now, if you're interested in registering for ASU's AI-Enabled

Supply Chain Strategy Executive Program this October, we have the details

in the show notes, including a nifty discount code for DataBlade listeners.

That's right.

If you're a supply chain leader in your org, this is a great opportunity to get

ahead with frameworks for evaluating AI opportunities, get a clear read on where

companies are getting genuine competitive advantage from AI, and be in a room

full of peers working the same problems.

Seats are going fast, so don't wait.

Register using the link and ID Code in the show notes and currently on screen right

here for $1,500 off the cost to register.

Now let's get into it for today's topic, procurement, and

welcome Professor Brett Duarte

Professor Brett Duarte, welcome to the Data Blades podcast.

It is great to have you with us

Thank you.

Good to be here

We've got three episodes to work through together, so why don't

we use this one to kinda set the stage, set the table a little bit before we get

deep into our procurement topic today.

And, let's just start w- with you.

Tell our listeners, who you are and the role you play at ASU

yeah.

my name is Brett Duart.

I'm a clinical associate professor in the NASPO Department

of Supply Chain Management.

I'm also the co-director for our business analytics program, and more recently,

I'm the assistant chair for graduate programs in the supply chain department

It's a good thing you got lots of time on your hands.

Over the past couple of years, we've seen organizations across nearly every industry

make significant investments in AI.

The success stories are many, but some are quite not so much.

What are you hearing about the biggest challenges that supply

chain leaders are facing?

Yeah, that's a, that's an interesting question because

many companies have invested a lot of money in AI, and there are

definitely a lot of success stories.

In fact, I think there's a couple of articles from the Boston Consulting

Group, somewhere in the middle of last year, and in terms of AI, last

year is already a long way back.

But, they talk about how many AI leaders seeing two X revenue growth,

in specific projects and also seeing, forty to fifty percent cost reduction.

so lots of value in pilots, but what's really interesting is there was another

article out of the McKinsey Group, later that year, which essentially

said that, while these companies have adopted AI, still less than one

percent of organizations that have really seen enterprise-wide impact.

and so the question is, where's the gap, right?

And, is it really an execution thing?

Is it a technology piece?

and so what we've done is we've spent a lot of time speaking to

supply chain leaders to get a sense of, "Hey, where can you help?"

"Where are some of the challenges?" And the, the messages that we're

hearing are somewhat the same, right?

One is infrastructure and readiness of infrastructure.

typical infrastructure tends to be very siloed, data

quality issues, fragmentation.

in many cases, AI technology is slapped on the legacy systems, which

really makes that end-to-end view of supply chains really hard to see.

Okay?

The other thing is really, adoption, right?

employees leaning in and, taking advantage of some of these tools and technologies.

and so if companies can really focus on these two critical things, the

IT infrastructure, really deploying, having a good deployment strategy

and getting their employees to learn how to use AI, think they can move

that needle from hype to value.

Yeah.

what are some of the things that, the Department of Supply Chain Management

is doing to really support and kind of partner with the industry around all

of these challenges you just listed?

Yeah, so like any education institute, ASU really has,

a mission of three critical things, and that's also pertinent to the NASPO

Department of Supply Chain Management.

The first being, education, right?

How do we train the next generation of leaders?

How do we infuse AI into our classes?

how do we, provide executive education?

We have some of these work sessions that we'll talk about.

How do we provide certificates for people that have not been in school

for a while but can quickly learn some of the tools and technologies?

But there's also an element of research, right?

and part of what we have done is we've recently created a AI lab with

the vision of being a leading hub for AI-driven supply chain research.

In AI world, there's this notion of creating a digital twin, and that's

what we're hoping to do, is to build a digital twin, which is essentially,

provides us, if you will, with this capability of having a living

laboratory for AI experimentation.

So we can come in, we can test agents, we can test different algorithms.

it really provides a great platform for experiential learning, because

what we can do is we can simulate a disruption, and then get a sense how,

how-- of how AI, especially AI agents, react to some of those disruptions,

and also hopefully train some of our students to work with these agents,

to manage some of those disruptions.

So I think this is, something that's really key from a research perspective.

And of course, we wanna be able to partner with the industry.

I there's the education piece, there's the research piece, but of course, we

wanna partner with industry, partner with companies, partner with the

community, to be able to hopefully utilize AI, to improve overall efficiency,

effectiveness, just quality of life.

that's great.

can you share a little bit more about the three-day AI enablement

supply chain strategy program that's scheduled for what?

October 5th through 7th.

That's correct,

What should we expect?

Yes, it's a three-day intensive, October 5th to the 7th.

it really is for senior leaders, executives, strategists, anyone

in the supply chain space, that's really looking to explore how AI

can transform a supply chain, like operations, procurement, logistics.

and quite honestly, the way it's designed is there's really a

little something for everyone.

So if you're completely new to AI, and you've never used it in your organization,

this is an awesome, opportunity to kinda get in some hands-on exercises on how

to build agents, how to use them to assist, to automate certain activities.

On the other hand, if you've spent quite a bit of time in AI, there's

something there for you as well.

We'll talk about infrastructure, deployment strategies, return on

investment, bringing in industry experts so you don't have to spend

three days listening to boring faculty.

we wanna be able to kinda pull in some industry experts, get a sense

of what they're seeing, and really provide this forum for dialogue, right?

what is your company doing?

what is our company doing?

what are the differences?

What can we share?

So it's really meant to be very interactive.

The three-day intensive is essentially divided into, three components, okay?

The first one focuses on the front end of the supply chain,

essentially procurement and sourcing.

lot of use cases on how AI can help play a role in that space.

Day two is on operations.

And so we'll talk about things like forecasting and inventory management.

And then leading into day three, we'll wrap up the back end of the

supply chain, focus on logistics, but then also bring in, topics like, AI

infrastructure, governance, business cases, how do you benchmark critical

success factors, deployment strategies.

So kinda rounding it all up together.

So quite honestly, a little bit of something for everyone.

Definitely does sound like it.

you've got really a lot of great material packed into that.

and then I think you even have some, some of the industry folks that you

have coming, if I understand the agenda, have actually been on our show before.

As I know Dave Finnigan is one of the folks who's coming in,

so a quick shout-out to Dave.

so you've mentioned a couple of times, you talked about agentic

AI and AI agents, and how that's a core piece of what you're gonna be

covering in the executive program.

And we've talked about agentic AI before on, on our show and h- a

number of topics related to that.

But just in case for anybody listening who's not as familiar as, maybe they'd

like to be with agentic AI in general, can you explain agents for us and how that's

s- such a relevant topic, how they might be used in supply chain management, or how

a practitioner would engage with agents?

Oh, yeah, absolutely.

agents, really the best way to describe them is, the digital workforce.

Okay?

They have, this ability to work, autonomously.

they can plan, they can reason, they can act, they can learn.

they can pull data out of different systems.

They can, evaluate it.

They can find ways to make decisions and learn from their mistakes.

from a supply chain standpoint, that really opens a whole spectrum

of use cases because now you can have these digital workers place

orders, they can reroute shipments, they can flag certain suppliers.

and of course, they do come with different levels of capability, right?

From the most basic ones, which perhaps can, really support you as

far as informing and assisting you with day-to-day activities, to, uh,

to, uh, moving up that maturity curve, if you will, with automation, where

agents can really automate workflows.

Essentially what would happen is you give this agent a goal, and it's able

to go through some multi-step reasoning.

It's got some memory and context, it's got some self-correction, and it's able

to execute on certain tasks, on its own.

So the whole idea is to kinda evolve from having a human with an AI as an assistant,

to this notion of becoming what we would call a frontier firm, where you now

have a hu- that sets the goal, but the agent essentially goes in there, works

the processes, creates the workflows, and then comes back and reports to you.

So this work session is really geared towards, focusing on agentic AI.

Uh, we wanna be able to, reach people that have never created an

agent, so there's a lot of hands-on experience on building an agent.

it's gonna be great because everyone's agent may be a little bit different, and

so we can kinda learn as we move along.

but then there's also the element of how do you now bring

all these agents together?

part of

Right

done, we've come up with this, curated end-to-end

case of a supply chain.

So we've essentially built data, that repre- represents a company, and

you've got, purchase orders and sales orders and, you've got demand, and

you've got logistics and distribution.

and what we're hoping to do is use this case as a foundation for learning.

So on day one, you build an agent to help you with procurement and

sourcing and contract management.

on day two, you build a agent to help with forecasting and inventory management.

And I'll talk a little bit more about what each of those do.

and then on day three, you build an agent to help you with logistics, right?

So you've got this small army of agents, and essentially what we then do is, we

simulate a disruption, and we're hoping that these agents can kind communicate

with each other, and solve some… or make some decisions that impact end-to-end

supply chain, operations, right?

the notion is if an executive is able to build an agent in three days if

he's never done before, then hopefully they're inspired to take some of

these learnings back to their company

Back with them,

exactly,

of the team, yeah.

it, there's no reason why you

Mm-hmm

be able to do it."

So on the p- procurement side, what would a specialist expect?

Do you have any use cases for that?

yeah, absolutely.

Quite honestly, Casey, each of these verticals we could

probably spend month, you

know,

Right

different topics- Yeah right?

Yeah.

The supply chain is so

Right

Uh, to try and squeeze all that information, into a, a couple

of hours is, definitely a challenge.

So what we've tried to do is, we're kinda looking at the

procurement and sourcing, will.

Think, typically someone in procurement or sourcing has certain tasks that are

very tactical in nature, looking at identifying suppliers, creating purchase

orders, looking at supplier performance.

But then there's also strategic elements that they get involved in, spend

analytics, category management, contract management, supplier risk management.

and so what we've done is, through our case, we will build a procurement agent,

that will be able to pull data out of this case that we've created and come up

with industry insight, maybe create some predictions, look at external markers.

maybe there's a tariff that's, imposed on a particular location, particular count.

How does that impact pricing?

Looking at both structured and unstructured data, pulling in sentiment

analysis because that's very important to get a read from maybe news articles,

on what's going on in the space and how does that impact, purchasing.

and then also looking at automating some of that communication between suppliers.

Instead of, pulling out an email and into your ERP system, pulling

information, we wanna find ways to automate that communication, really make

it more efficient for the user, right?

Now- The second piece of the whole procurement, space, centered

around contract management.

And I think smart contracts, is really a space where there's so much

of efficiencies to be gained, right?

because what a contract manager does essentially is reading

through multiple contracts.

You're looking at terms and conditions.

You're extracting performance.

You're looking at risk.

You're notifying suppliers.

you're updating the contract.

Lots of manual work.

And what we're hoping to do is to demonstrate how a contract

agent can really automate a lot of that functionality, right?

essentially looking at your contracts, pulling up ones, which are, have expired

or perhaps which are ready for renewal.

Evaluating those contracts against sales data or, purchase orders to get a

sense of which contracts are violating some of the terms and conditions.

things like, on time delivery, lead time, quality related issues.

and then setting up, meetings with your suppliers to say, "Hey, guys, it's time

to renegotiate contract based on here is

right?

over the years.

we might need to talk about renegotiating contracts." And then

finally, I think what's really cool is, this, this notion of looking at

your suppliers, and then coming up with ways to kinda mitigate risk.

So we'll talk about things like the Kraljic, matrix evaluation, how you

come up with supplier scorecards.

and then really automating that communication, between the

suppliers and, and the company.

It sounds like quite a lot to pack into that one day in the class.

I think for most procurement specialists, that's gonna be an exciting, amount

of material to cover to really help them optimize their day-to-day.

Uh, absolutely.

It's, it's a packed session, but, we're really looking at making

it very interactive, right?

and

Right

y- engaging with small teams, having people work

together, is the way to go about it.

We find that, learning from each other and kinda getting a sense of, what are

some of the best practices the way to go.

But day one is super exciting because it really sets the stage,

for what's to come, down the road.

Yeah, and I think there's really a good lot of benefit to

be had because you'll have people from different industries, so there should

be a lot of kind of car- cross-industry thought, I think, that everyone would

learn from as well, even though they're probably experiencing similar problems,

just maybe attacking them a little bit differently within each industry

a- absolutely.

and then there's different levels of, maturity of AI within

their organization, right?

Right.

are just getting started and don't know where to start.

and others have, as I've mentioned, been on that path for a while, but

are looking at ways to fine-tune their strategy, and their roadmap

Well, I think that's it for part one today.

Brett, thanks for sharing with us, covering day one of the program.

we're not done with you yet, though.

Yeah, not even close.

Next episode, we move downstream into operations, forecasting inventory, and

the gap between a model that predicts well and a business that acts on it

If procurement is where you decide what to buy,

operations is where you find out whether you got it right.

So that'll be part two, and that is coming up next week

Follow Data Blades so you don't miss it.

And if this series has you curious about the full three-day program

at ASU, a quick reminder that the registration link and our listener

discount code is in the show notes.

Ricardo, this episode is a wrap

See you next episode

Love this episode?

Drop us a five-star rating and review on Apple Podcasts, Spotify, or Goodpods.

And if you're watching on YouTube, like and subscribe before you go.

I'm Casey Golden.

Follow us on LinkedIn, Bluesky, Threads, and Instagram,

and subscribe to our Substack for highlights and bonus content.

For transcripts and guest info, visit retailrazor.com.

I'm Ricardo Belmar.

Thanks for joining us on the Retail Razor Data Blades, part

of the Retail Razor podcast network.

Until next time, stay sharp, be data-driven, and harness AI.

This is the Retail Razor Data Blades!