Why the Biggest AI Transformation in Retail Is Not What Customers See

Ghazanfar Riaz Sheik
If the retailers are still waiting on to see hey everything is going to be I would say like vetted out and then we're going to adapt to it then I think it is going to be too late.

Samuel
Ghazanfar is senior vice president of the global digital solutions group for the Microsoft business unit of Visionet with 20 years shipping enterprise technology. He's now deploying AI into some of the biggest retailers in North America.

Ghazanfar Riaz Sheik
The frontline AI is not just a chatbot, but it is a next action engine for a person running that store. Everywhere that you go, you expect the same kind of level of service and if you don't get that, then that's an issue. That's where you would lose the customer. The real risk is falling behind in organizational learning because every organization is different. Start with the daily friction the store has already learned to tolerate. I think that is the most critical thing that they should focus on. The frontline workers will not be replaced by an AI agent. That is for sure.

Samuel
What if the biggest gap in retail isn't between you and the product you want, but between the associate helping you and the systems they can't access? What happens when the associate knows less than the customer standing in front of them? That's the gap Ghazanfar Riaz spends his days closing. Ghazanfar is senior vice president of the global digital solutions group for the Microsoft business unit at Visionet. With 20 years shipping enterprise technology, he's now deploying AI into some of the biggest retailers in North America. Restocking shelves, running the till, checking if this size is in the back.

Millions of people do that job every single day, and every one of us deals with one every week. This episode is about how AI is changing that job. Before we jump in, I have a small favor to ask. If you're getting value from this show, hitting subscribe, dropping a comment, or leaving a like is one of the best ways you can support me. It helps more people find this conversation and honestly it means the world. Thank you. Now let's jump in. Ghazanfar, welcome to the AI Frontier Playbook. Ghazanfar, you're the senior vice president of the global digital solutions group for the Microsoft business unit at Visionet.

You've been deploying AI into some of North America's biggest retailers across estates with like thousands of stores. So today I like to talk about how AI is affecting frontline workers in retail. I'm not talking about chatbot on a website. I'm I'm talking about the person restocking the shelf, running the till, checking if that size is in the back. I'm talking millions of people. Probably one of the most common job there is you've already put this technology in their hands.

So I like to start there because when I air AI in retail, I'm mostly thinking about shopping assistance. we've all used one of the chatbot on the different biggest retailer out there. so I'm thinking about more about that, not the persons talking the shelf. So today I like to understand what does AI for a frontline worker actually look like in practice? what kind of use cases are we talking about here?

Ghazanfar Riaz Sheik
Yeah, thank you.

So, thank you for having me on the podcast Sam first of all and I'm excited to be here looking forward to our conversation and a great I would say topic from a frontline worker perspective because when most people think about AI in retail they picture a very shy digital assistant helping the customer shop right but the bigger transformation I would say is on the floor and that's where AI working along alongside with the frontline workers within the store floor is an epic I would say change across the board and helping them see what needs attention what kind how to make the right kind of decisions and then acting before a sale or a customer is lost that is much more critical these days for retailers and if I want to bring this up from a life through the typical day in a store for a frontline worker or a store manager perspective.

That's where the a lot of shift is actually happening, right? So, think through from an angle where an associate actually starts their day. they begin a shift instead of checking emails or printing instructions from the day before or multiple going through multiple applications a Copilot assistant can actually provide a short brief to them for starting the day. So it can actually tell them that hey these are the five priorities that you need to concentrate on. these are out of the those five priorities. Two promotional products need repl orders are to be packed and then shipped. Right?

And then there is one or two customer issues that that I have worked through. Now I have to contact the customer and get that thing resolved. Right?

So these are the kind of things that actually happen from a day-to-day perspective from a from the frontline worker perspective and then later in the day AI can actually help or any Copilot agent can actually help detect that what are the item that are being promoted on the website as well as for the storefronts and then if they are missing from the shelf if it is if it is then it would automatically check on the inventory confirm that hey out of these many different SKUs. there are 12 units that are available within the back room and then creates a repment task so that the right kind of associate can be associated to that particular task and then they can go fill that up so that any customer who walks in would find that appropriate promoted product and into it.

So this is where I would say AI can also help from a lot of visibility aspects a lot more on the productivity side and then one important I would say another use case that I can think of is an employee who is who's new to the store who's new to that job and then don't know the SOPs of that store they don't know the complexity or the processes in in all in all and they heavily rely on others to basically guide them on that.

That's where an associate might be preparing a pickup order or they might be looking through certain return u return processes and that's where the guidance coming into the play from the AI side right so they would be able to complete all those kind of things by utilizing AI rather than going and asking questions again and again from either the manager or their senior associates.

So these are the kind of use cases that are there predominantly within within the day-to-day thing but as consultants I would say our job is to map those components together around the actual frontline journey that's where we bring that connectivity end to end and then does not overload those frontline workers with so many different applications but rather focus on the things that they need to get done the create the why for the customer experience create the focus on the productivity aspect and that's where the frontline AI is I would say not just a chatbot but it is a next action engine for a person running that store.

Samuel
It's true that as consumers, when we're going into a store, we don't think about all this backend that's happening, all those applications that the associate needs to learn and work in. From what I'm hearing from your first use case, it's mainly becoming more proactive than reactive, right? It's giving access to the insights that the associate needs before they have to think about it, before looking at the shelf and seeing that it's empty and needs to be restocked. Now you can use AI to help the associate plan their whole day, telling them, hey, you will probably miss some of this item on this shelf and you need to think about it.

And your second use case as well around training, I will assume that it's a business where there's a lot of turnover, and having to train everyone on all the procedures is time-consuming and you need senior employees to train the new employees. So those are really great use cases that let you be more proactive and, correct me if I'm wrong, give more autonomy to the associate in the store.

Ghazanfar Riaz Sheik
Absolutely.

And that's the focus, right? Because as customer wait time increases, the CSAT score goes down. What we have seen is that the more the customer waits for an answer or for a product, the more likely you are to lose that sale, and your conversion and sales rates go down. So it's better to have those insights available up front. From a customer angle, I can see that they know what they are selling and are able to help me with everything. If you're able to help the customer find a better product, help them with that discovery, and understand their intent, I think that is where the conversion rate will definitely go up.

Samuel
It's interesting you're mentioning that. I was shopping for a backpack recently and went to one of the big retailers out there. I asked a question and got a totally bad answer about the product. I ended up using Copilot and asking the question myself because I had a doubt about what the associate was telling me. As a consumer, I would have been happier if they had picked up their phone or any device from the store, asked the question, validated it, and given me the right information instead of guessing and finally giving me the wrong information.

So the customer experience was very bad, and that's the kind of experience you can enhance using those tools. This is incredibly powerful for a consumer as well.

Ghazanfar Riaz Sheik
And with AI, customer expectations are going up every day because they've started to use AI, ask questions, and get prompt answers. A cadence is basically being set, or an expectation is being set for that portion of the mix. Everywhere you go, you expect the same level of service, and if you don't get that, then that's an issue. That's where you lose the customer.

Samuel
Now we're all used to using AI in our daily lives, with ChatGPT, Gemini, Copilot, or whatever you're using. For consumers, it's easy because it's connected to the web. It's connected to its own knowledge source. But when we're talking about an organization, it's a bit more complicated because there's technology already in place. There's data that lives in different systems. There are people who need to use those technologies. There's an adoption curve. There's training to be done.

So what's the actual bottleneck you keeping when dealing with retailer? Is that technology, data, people or a mix of everything?

Ghazanfar Riaz Sheik
Yeah, that's a great question for organizations like every organization is different at a very operates very differently, right?

And if we look through that lens, it's to be very honest a million-dollar question. it's not a just a cookie cutter answer but to the experience that I've had so far I would say I would say the focus is on people right why because like more specifically I would say the people specifically from the frontline perspective right because that's where the trust and the operational ownerships comes in into the play right a frontline worker makes decisions in real time often while a customer is waiting thing and if in case the AI provides an inaccurate answers or requests too many steps they would basically ditch it right they won't be able to they won't use it because they are in a very fastm moving thing and they have they do not have time to experiment with the technology during a busy shift and that's where I think where the AI can be can be helpful but at the same time if the SOPs are not clear if how the AI is set up for them or any if I set up a Copilot agent and it is responding with incorrect answers or the SOPs that have not been vetted out then that I think is a problem of course like there might be many other problems connectivity multiple different systems getting the insight from all across the board but I think the best capability aspect all boils down to just one factor which is people which is the frontline workers right?

If they are not utilizing the technology right or if they are not involved up front in into designing or solutioning of that particular AI solution that it it won't it won't create a difference for them and they won't be using it end to end.

So that's the reason why we start with a focus on the frontline workers, looking through their processes and identifying who the business owner is for that particular portion of the mix, where the source of truth is, who is involved from a storefront perspective, and who is going to be the change champion and take ownership. This needs to be designed in the right way so employees can use it to its full extent and be more productive. I think that is the most critical thing to understand. Accuracy and first-time resolution all boil down to how those things are set up for the frontline workers.

Samuel
So yeah, it's a bit like for office workers, right? If you give them a tool and they don't use it, you won't see any improvement, right? If you want a collaboration between AI and people, you need to train them first. I'm keep seeing that with my my own customers, right? office worker, you give them access to Copilot or Cowork or scout or whatever technology, but if they're not trained on how to use it and if they don't incorporate it into their daily workflow, it won't have the expected impact. So I assume it's just the same with field worker, right?

Ghazanfar Riaz Sheik
Absolutely. In our experience, we have seen adoption go up exponentially because the outcomes of utilizing AI are faster and more consistent, frontline workers see fewer operational errors, and associates can focus more time on the customer. That is the reason why we are seeing that adoption. When they come in on a daily basis, look through different things, and get those kinds of insights through Copilot, it adds to their productivity and they're loving it. But again, it all boils down to people, change management, and adoption.

Samuel
When you walk into an account and start scoping that kind of project, what tells you in the first few weeks whether they're actually ready versus whether this is a project that is going to stall?

Ghazanfar Riaz Sheik
Well, I would say that's a tricky question because you can tell by a few traits, but again, organizations differ in adoption from an AI perspective.

Yeah. some organizations think that they know what they want to do from an AI perspective what that what that would look like for them. What kind of vision do they have?

But once we go in and then spend few weeks with them they don't really have the clarity and then of course everyone is evolving right every organization is evolving towards AI everyone is looking towards doing some kind of an experiment and then learning lessons from it but our SAS is that when we go in u there are a few things that we basically look for and specifically what we look for is from a business ownership perspective what who is a business owner what are the measurable outcomes that they're looking for from a AI adaption perspective right that that's number one that is the most critical thing and then can can that particular retailer identify have identified at least few high frequency problems instead of gazillion AI ideas that they might have right so the focus should be on a limited set of use cases particular ularly related to a certain department.

It can be the stores, it can be the accounting, it can be the HR, it can be the supply chain aspects or the production lines or maybe the frontline shops, it can be anything but there has to be certain high frequency problem that they want to solve for. And then the respective ROI part, right?

I think that is a very very critical thing where you might want to implement AI anywhere and everywhere but it is something that you need to think through from an ROI perspective because that what is going to cost you for if if it is not creating that kind of productivity aspect and into the mix right and in addition to it again like that that's from the process perspective or the milestones or the u key achievements or the vision that they want to achieve. But other than that, where does the data reside? Where do they even have that kind of data or the necessary data entities that would be required to up that particular AI part?

That is a very very critical thing. So storefronts or associates are they also involved in into the change management aspect because they are the process custodians right if I look through from a frontline perspective. So those are the kind of thing that we look through and then we want we would want to say that hey these are the kind of guiding principles where you need to think through from that angle and that's where the readiness of I would say the KPIs come into the play. So all the business outcomes are shorter from our development life cycle.

But who's going to be the owner, what it is going to do for them, and what kind of insights it's going to generate are the kinds of things that take a bit longer, and the organization should spend the right amount of time on them. That's just my two cents.

Samuel
So, start with a KPI or the outcome you're looking for instead of starting with the technology, right? Is there a common use case you keep seeing? I'm curious what kinds of use cases you're seeing retailers use right now.

Ghazanfar Riaz Sheik
Absolutely. From the use case perspective, a retailer we recently worked with came to us with a very interesting use case. They called out that their shelf exceptions were occurring approximately 100 times a week across thousands of stores, right?

And it currently takes on an average about 25 minutes to identify and correct that issue. And then only 70% are resolved within the same same shift. And then the VP of store operations they had specific KPI that they wanted to execute like they wanted to have availability and quality of of the SKUs that were available at the shelf.

They wanted to make sure that the frontline workers were were more accountable from a replenishment perspective. they wanted to make sure that they set the right kind of baseline from the SOP's perspective that whether if you're helping out a customer with a certain thing within a certain period of time they should be able to produce that those results right so those are the kind of I would say underlining use cases that company came forward with and that's where we actually help them u develop a system end to end where we always call out that from readiness perspective like they were ready with and with clear goals that hey this is what we want to achieve from a from from the utilization of AI and that's where they focus on the problem and then we go in and then provide the solution to that and then that creates a difference for those KPIs as well that's where they become more resilient that's how they would be able to bring them up to speed from a KPI perspective Yep.

Samuel
And as a consumer, what will I see different when I'm walking into a store that has been AI enabled? why should I care as a customer?

Ghazanfar Riaz Sheik
Oh, great question. I would say like u, from a customer experience perspective, what does a customer wants, right? they want to when whenever they are walking in into the store, what they're looking for is a frictionless experience across the board. whether they're going in, they're going in for a reason. They're looking for something. If somebody can help them out from that angle, if somebody can understand what they're looking for in a better way, articulate the problem statement much more clear in a clear sense to them, that would ultimately be the connecting bond between the customer as well as the frontline worker. Right?

So consider a customer looking for an item that is unavailable in the correct size or today an associate may go and check the shelves, search the back room, ask a manager and eventually tell the customer that hey we don't have that in stock. Either you can try it in on our website, order it and then it can be shipped to your home or things of that sort. Right?

But with the but the time that has already been spent by going through and looking through it that's where the frontline workers can actually create a difference right and by utilizing Copilot agent I would say assisted selling options as well as the crux of it the associate can actually see the inventory across nearby stores in that way they would be able to make recommendations or for a substitute they can make recommendation that hey this is something that is nearby, I can actually just click of a button, I can actually order it and then you'll get it at home or you can just pick it up on the way back to your home as well.

So those are the kind of things that they can actually use and on top while I have all those kind of insights if I for that inconvenience if I actually apply some kind of a promotion by utilizing by utilizing the AI aspects and into it and then placing that order that would actually create a wow factor for the customer. Right? So this is this is I think a great example that one of our one of our retailer customers actually implemented and has created a lot of loyalty as well as like appreciation from the customer side as well.

And I think the biggest use case I would say is returns right rather than sending the customer from the associate to the service desk or then to the manager an agent at the front end can actually interpret the the intent of the customer coming in and then it can actually help them out rather than them being in the line or looking through all those kind of things. So by utilizing whatever the order number that they had or the receipt, it can actually scan that and then quickly pull up information and it can be a seamless a seamless experience for a return site. Right?

So this is where I would say the the wait time goes drastically down. and one of our retail customers have actually implemented that and it it has significantly reduced the hours that were being spent on the return just on the return side. Right?

So this I would say increases the outcome not just on the customer satisfaction part but also helps the the same customer to buy more and more from that reh from that brand because the things and the processes are easier for the customer as well and customer may never see the AI but like they experience it as a better informed I would say store or the brand and then that's basically what it keeps its promises end to end.

Samuel
So, from a frontline worker perspective, if I'm using this use case, I'm starting my day and receiving an email on the device provided by my employer. I will assume that will provide me with insights into how to go about my day. Then, while I'm dealing with a customer, using the same device I will have a chatbot or an agent that will help me upsell, find information about the customer, check inventory, and check for promotions. Am I right?

Ghazanfar Riaz Sheik
Absolutely. I'll give you an example. With one of our retailers where we deployed Dynamics 365, we have seen the experience grow exponentially for frontline workers.

So for example a customer comes in and then brings in a a product they would be able to see what previous transactions that they have done what are the thing that they liked right what are the kind of products that they are product line that they are more aligned towards right and then that would actually give them insights in into having a icebreaker conversation with them right or what kind of discussions can I do with them or what kind of product lines can I actually pitch while they are at this at the store and then checking out, right? that would actually create a mesmerizing experience for the user that hey, yeah, absolutely.

This would actually go very well with what you bought last time, right? And this is the connectivity that customers are looking for with that particular brand. I would say that's where we have seen increased revenue, stronger loyalty, and a more consistent experience across the board when they start to utilize these kinds of insights from a POS perspective.

Samuel
So this is basically the experience that we're having on the web, now transposed into a more traditional brick-and-mortar retail store. This is really interesting. I'll be honest, I'm mostly buying online because of this experience.

If I could have the same experience in person, I would totally go back to the more traditional shopping experience.

Ghazanfar Riaz Sheik
Absolutely. I think it's human nature to want that connection. One-to-one, person-to-person connection is never going to go away. So elevating that experience is key for retailers. Giving more insights to the customer, as well as to frontline workers especially, is going to be a game changer for a lot of retailers going forward. That's where the productivity, resilience, and revenue growth they're looking for will come through.

Samuel
You're in these scoping conversations, I think, before most people even know they're happening.

Samuel
So what do you think a retailer actually risk by waiting another year? Is it going to lose ground to competitors or is more that the technical depth of catching up will better get worse or a mix of of those answer or something else?

Ghazanfar Riaz Sheik
Yeah, we we've seen mixed reviews on that but I think the biggest risk for from a retailer perspective is not missing one generation of AI technology, right? The model will continue to improve. The real risk is falling behind in organ or organizational learning because every organization is different.

Every organization works in a very different manner and retailers deploying AI now are learning how to connect their products, inventory, customer feedback, insights, workforce data, all those kind of things is something that is helping them to govern the the agents that the functionality of the agents, how to redesign their processes or re-engineer their processes and how to earn the employee trust.

I think each department creates another feedback loop and that is where if the retailers are not adapting to AI early on then I think like they're going to be way far behind from other ones and I I can give you an example right like one of the biggest I would say shoe manufacturing company in the in the whole world we've been working with them for quite some time and then they they basically begin with a knowledge agent, right?

And they basically connected it to ask certain questions internally from from the IT perspective and then they basically added more and more use cases. task management, inventory exceptions, assisted selling was the last one that they started with where they were actually selling hey based out of your based out of your shoe size as well as how your foot is basically I would say like measured and then how if your foot alignment all based out of that like it was actually giving insight.

So that was something that after a year it was not just simply implementing few of the use cases it actually created a lot more insights for them from a product product research and research and R\&D perspective right in that way they gathered a lot more insights from the customer that were actually buying those kind of products and why they were buying those kind of products right and at the same time they were these use cases were actually collecting a lot more data from the storefront or from the stores like where or what kind of processes can be improved where is the lag what kind of trainings are now needed for the frontline workers what are the things that we can provide to them to elevate not just the customer experience but also increase their productivity aspects right so these are the kind of things I would say that are very very critical and then if the retailers are still waiting on to see hey everything is going to be I would say like u vetted out and then we're going to adapt to it then I think it is going to be too late.

So the technical debt or those are the kind of things do matter yes but the greater risk is operational model debt. that is where if they don't adapt to it they won't be able to sustain a com competition against their their competitors right so I think it is very very critical for retailers to adapt to AI and then look through and then experiment and in that way they would actually learn a lot more about their own organization.

Samuel
I think a lot of organizations, not just retailers, underestimate the adoption time and the learning that needs to be done.

Ghazanfar Riaz Sheik
Yeah, I totally agree. The competitive gap becomes so huge because the time from idea all the way to production, then enablement, change management, and adoption from a store or customer perspective is huge. That has a bigger impact on their margins as well as revenue growth.

So I think those are the kinds of things that organizations should definitely think through, and then adapt to the AI changes that are happening in the world.

Samuel
We're almost at the end of our time, but before I let you go, I'd like to hear from you. Given everything you've seen deployed on the floor, what's one practical thing a retailer, or even just a store manager, could start doing with AI tomorrow that will actually move the needle for their frontline team? I'm not talking about a long project, right?

I'm talking about something that's available today to them.

Ghazanfar Riaz Sheik
Again, I think they should start with an AI-powered shift huddle using their own retail environment. Store managers can combine yesterday's sales, today's promotions, staffing constraints, and inventory gaps. Those are the kinds of small things they should start with. AI then creates a briefing, as I said earlier. From a shift briefing perspective, those are the top things they should at least start with, and that is where they will be able to identify a lot more use cases that are aligned to the KPIs they have within the store.

So for example I would say like manager would actually be preparing a lot more schedules replenishment task inventory checkups all those kind of things is something that the store managers actually prepare for a day before and then for the next day they basically define it and then people come in look through those notes emails or charts on a regular basis right what if the AI can actually help with that portion of the mix that's where it would reduce the shift hurdle duration.

It would basically be able to complete the critical tasks and then the completion rate would actually go up on that part, right? and then overdue tasks or things that are missing their SLAs or if we have inventory issues. Those are the kind of things that you can get a lot more insights up front and then can actually place orders or work on those kind of feedback aspects that I think are the most critical things where the business outcome is better from a daily execution perspective and then without waiting for a larger transformation program.

So this is where I would say that start with the daily friction the store has already learned to tolerate. I think that is the most critical thing that they should focus on and this is something that can be addressed with out of the box tools that most organization have right now.

Samuel
I'll take Copilot example or Cowork where you can really just export data from your existing system running to Cowork or Copilot or any other AI tools you're using in your organization and then get get insight very quickly and be able to adapt without having to wait for your IT team or organization to go through this whole AI transformation.

Ghazanfar Riaz Sheik
Absolutely. And and that's where you don't really need to do any coding. the Copilot the Copilot actually would actually be able to help you with that part.

So for example a manager should capture at least 10 to 12 questions that the associate asked mo most often right like that was one of the use cases that one of our retail customer actually brought in and then say hey I want to help my manager adapt to it and without doing anything I just want to show them that hey just by configuring this agent it would actually help you with that part and these are the 10 question FAQs that mostly like your associates are going to ask you and those are the questions that were put in into that agent. And then it is now actually helping out the the associates the store associates.

So they basically instead of going to the manager they just ask Copilot about that question and that query and it actually gives them the complete SOP details on that. So it was without coding available to them where they can leverage that part. Very simple use case but very powerful.

Samuel
Last question. Zooming out 10 years from now what does the frontline retail job look like? Is it even recognizable or does AI change what that role means entirely?

Ghazanfar Riaz Sheik
Wow. you we can bet on it on KCI and then see in like 10 years if you're going to win that or not. But u like jokes apart the I I think the job will still be recognizable because retail remains a human environment.

I think that that's that is not going away at all. customers will continue to value judgment. they look for connectivity with what they are buying and then that's the reason why empathy and personal service are going to be key aspects for retailers right it would continue to be the most critical thing but there has to be a balance right there has to be a balance from the change considerably and that's where I would say the imagine I would say the store associate begin their day with a team of digital agents and then one monitoring shell, one prioritizing orders, one identifying loyalty opportunities, one watching the operational risk of how things are happening and that's how it would basically help them out and then the frontline workers or the store manager would actually become an orchestrator of people AI agents and store economics end to end that I think is going to be a key and that is where I would say the criticality would come into the play because the KVIS would change a bit from a feature perspective.

The routine work of how it is automated are going to be a key aspects. how well their employees are trained from an SOP perspective. Those are the kind of things that are going to be much more important. But the business outcome is a much more productive store, right? With a smaller administrative burden. that I think is going to be a key aspect when it comes to how I look through from from that role in the next 10 years.

Samuel
Honestly, I can't wait to have the same experience in traditional retail store than I have online. So, I'm looking I'm looking forward for for those change and I think it will be even more enjoyable for people working on the floor.

Ghazanfar Riaz Sheik
Yeah. And and like one one myth that is going to debunked is that the frontline workers will not be replaced by an AI agent. that is for sure. there's a lot of skepticism about that and that is not going away. they will become I would say the person directing a team of agents while owning the customer relationship. So the most critical thing or the valuable human role comes in into the play is the customer relationship.

Samuel
Yeah, I I totally agree and that's that's why I'm still going to to physical stores to have this relationship. So I totally agree with you. I don't think that people will be replaced by an AI soon and I hope it will not. It will just empower them and give them more autonomy and let them enjoy what they like to do which is probably the relationship part of their job.

Ghazanfar Riaz Sheik
Absolutely.

Samuel
Ghazanfar, thank you so much for your time today. was a very enjoyable conversation. learned a lot about where AI or retail is going using AI and I think we have a good framework here to for retailer to start implementing AI in their own organization. So thank you so much for your time.

Ghazanfar Riaz Sheik
Absolutely pleasure to be here and thank you for having me.

Samuel
Thank you. Have a great day.

Ghazanfar Riaz Sheik
You too. Take care. Bye. Thanks.

Samuel
Ghazanfar kept coming back to the same point. Not the technology, not the data, but whether someone on the ground actually owns the problem and can name it clearly. That's the gap behind every associate who couldn't tell a customer if his size is in stock. And it's a gap AI is starting to close. Three things I want you to walk away with from this episode. First, before scoping any AI project, ask how clearly the team can define their own problem. If they're still speaking in generalities about performance instead of naming the specific bottleneck and who owns it, project is going to stall.

Second, put an AI Copilot in the hands of frontline associates so they can pull real-time inventory substitutes or ship-to-home options the moment a customer asks instead of walking to back room or guessing. That's the gap between a last sale and a converted one. Third, don't wait for the technology to match before you start. The real risk of waiting a year isn't falling behind on tools. It's losing a year of learning how your own data, your agents, and your store processes actually work together.

If you got value from this one, subscribe to the AI Frontier Playbook whenever you listen and sign up for the AI Frontier Playbook newsletter to stay sharp between episodes. Thank you so much for listening and I'll see you in the next one. See you.