On this episode of the Futurum Tech Webcast – Interview Series, host Daniel Newman welcomes Sameer Verma, Head of Product – Dynamics 365 at Microsoft for a conversation on the launch of new Copilot and demand planning capabilities in Dynamics 365 Supply Chain Management. They discuss how organizations can empower their employees with natural language-based conversational help, enhanced inventory visibilities, and improved accuracy, all with a more flexible, simplified, and intuitive user experience.
Their discussion covers:
- The current state of supply chains in a post-covid world and how the new state of work impacts how companies transform their supply chains to meet customers’ needs and keep employees happy
- Microsoft’s lessons from investments made in the past year and what future investments in Copilot might bring
- An overview of the new demand planning capabilities and the role of Copilot within Dynamics 365 Supply Chain Management
- How companies with “best-of-breed” demand planning systems today can benefit from the new demand planning capabilities in Dynamics 365
If you are interested in learning more, you can also download a copy of our research report, Enhancing Demand Planning with Microsoft Dynamics 365, today.
Watch the video below, and be sure to subscribe to our YouTube channel, so you never miss an episode.
Listen to the audio here:
Or grab the audio on your streaming platform of choice here:
Disclaimer: The Futurum Tech Webcast is for information and entertainment purposes only. Over the course of this webcast, we may talk about companies that are publicly traded, and we may even reference that fact and their equity share price, but please do not take anything that we say as a recommendation about what you should do with your investment dollars. We are not investment advisors, and we ask that you do not treat us as such.
Transcript:
Daniel Newman: Hey, everyone, welcome to another episode of the Futurum Tech podcast. I’m Daniel Newman, CEO of The Futurum Group. Very excited for this episode of the Futurum Tech Podcast interview edition. We’re going to be talking about a little bit of AI, some Copilots. We’re going to talk about what’s going on in ERP, demand planning, and start with a topic that’s near and dear to my heart, which is supply chain.
During the past few years, the supply chain became more and more visible to everybody in the world. We saw shortages in the semiconductor space that led to automobiles and smartphones and PCs becoming unavailable upon unparalleled demand. But we also learned a lot about what happens when supply chains get broken and when we don’t have the visibility or we don’t fully understand customer demand or changing trends. And with all the data and all the technology that’s out there, we have an opportunity to do this better. And I’ve spent the last few years looking across the industry at different companies that are building technologies and tools and utilizing data to help companies do this better. And one of the solutions that has come to my attention has been Microsoft and what the company has done with its demand planning and supply chain capabilities, and of course, the utilization of its Copilot. Its very exciting new generative AI technology and how this could be the beginning of a revolution that could remove some of the friction that’s caused some of the problems over the past, and by the way, some of the problems that still remain. So with that, I want to invite Sameer Verma, the Head of Product at Microsoft to join me here on the show to talk a little bit about what’s going on in supply chains, demand planning, AI, and so much more. Sameer, welcome to the Futurum Tech podcast.
Sameer Verma: Thank you, Daniel. Pleasure to be here with you.
Daniel Newman: So you heard me do my little spiel in the beginning about what I saw. I will tell you this, Sameer, I did 50 television appearances inside of 12 months about supply chains. A lot of them were about chips, but sometimes it was about cars or devices that people could not get their stuff. I mean, heck, it was toilet paper. We remember that. And while in some ways it feels like things are better, I would love to get your take on what supply chains look like following the events of the past three years and how you see this particular space.
Sameer Verma: Awesome, thanks, Daniel. I think, like you said earlier, the stability is definitely there. I think what’s changed dramatically for us, what we are seeing as vendors and also our customer situations, the whole evolution of technology. So with the coming of generative AI, our ability to serve the customers has become dramatically influenced by this technology. So in the past, customers had a lot of analytics dashboard reports, whereas now we are moving to this era where everything is you can talk to your system, get your data, be very, very real time. And for us, as Microsoft, what’s really helping is that we are able to orchestrate all the elements of our portfolio, whether it is on the productivity Office side, collaboration with Teams or the entire AI Copilot capabilities into the product. So the differentiation that we are able to bring through these technologies in our existing solutions is really the ability to orchestrate collaboration, drive those insights, which by the way were possible, but it required a lot of time to get to that state. And very often by the time you got to that state, the problem has shifted somewhere else. And we’ve seen that with people trying to build control towers command centers for every little thing. But by the time you really realized the full potential of it, the problem has shifted elsewhere. So for us, the key imperative is that, okay, customers, yes, there’s stability in the system, but there’s still a lot of ups and downs. So the question is how do we bring all these technologies into our offering and bring it to the customers where their workflows are, right? So instead of changing realms or starting yet around the project or implementation. So that’s how we are looking at this whole thing, and I’m happy to go into further details as we talk.
Daniel Newman: Yeah, I like what you said about the fact that this has been available because I think sometimes when a new technology gets introduced, Sameer, we act like, oh my gosh, this is the first time it’s ever been here. We’ve seen it with the last year, and Microsoft has made massive strides and been incredibly disruptive with its partnership with OpenAI. It really brought the ChatGPT product to market and made it very digestible. But we know that generative capabilities have existed for some time, and this is very evolutionary, but every so often the tech hits that tipping point and it becomes revolutionary. And so what you said there that was really important to me was making the technology easy, making it accessible, making it usable, because if you had enough data scientists and enough access and enough software, these kinds of solutions have been able to be cobbled together for some time. And that takes me to the Microsoft question. So I could look across the stack and I’d say there’s been companies for a long time that say we are the SCM companies, specialty supply chain. We do that software, we attach to a Microsoft ERP or we attach to an Oracle or you know what I’m saying, they attach to SAP. You are looking to simplify that and streamline that. If you could talk a little bit about how and why Microsoft saw this opportunity to really disrupt or really revolutionize the supply chain and really bring generative AI into this deep enterprise app layer, not just some consumer or productivity tool.
Sameer Verma: So I think that’s a great question. Probably there are three elements to this. So first of all, when we look at this whole capabilities of generative AI, which we call as Copilot here, it really is about augmenting humans. So it is not really about automating everything or making everything autonomous. It is really about if I’m a user of a software today, how do I dramatically change my relationship with the software? It’s almost like you think of Copilot as an agent that you hire and you put a lot of cognitive load onto the software versus the software telling you what to do. So that’s really the first principle, so to say. And what we’ve done really is if you look at any end-to-end business process that companies use, whether it’s source to pay, audit, cash, whatever it be, there are certain roles there. There are certain jobs to be done. And we are really using this framework to identify, okay, what is it that people are doing today and how can Copilot step in? And pretty much in some cases you don’t take away all those mundane tasks. Those are pretty much the low stake productivity skills as you called out.
The second piece is how do I offer the user, the insights, the blind spots, the next steps through this? This is actually pretty amazing. We’ve seen that, for example, in June, we released a procurement agent, Copilot. What it does is we have companies with whom we are working here who have like 85,000 changes in their confirmed purchase orders with the line items. Now in today’s world, any procurement agent needs to go through each and every line and then decide, okay, whether to approve or take a certain action. So with the Copilot that we release, for instance, it would tell you, okay, what are the line items which have zero impact on your downstream sales orders, production orders, or anything? So you can just go ahead and approve all of that, really look at capabilities, look at changes which have impact, right? And then you don’t parse those impact or what are the immediate ones, like what’s going to hit you first and what are the things that you need to do? So for example, the Copilot tells you, “Hey, here’s the supplier, probably this is what you want to address to them.” And this is where this whole orchestration of Outlook, Teams, Copilot comes together, where it prepares for you, okay, what is it that you need to potentially communicate to your supplier, brings all the context, and boom, you’re done. So what it has done to customers with whom we are working is taken away massive chunks of work, which will completely of no value because they had to be done, because companies need to be compliant and auditable. They have to go through all the changes in the purchase orders. But the reality is now with Copilot, you’re really focus on where there’s an impact.
So this is just one example, the other element that we are looking at is especially when you look at customers who are best of breed vendors who’ve been doing demand planning or supply chain planning as such. The way we are approaching this whole thing is enterprises, the whole definition of ERP supply chains has dramatically changed. In the past, ERPs were largely about four walls of business and whatever you do inside, now we talk about connected enterprises, collaborative planning, all of that, and you need systems which do that. You don’t need to stick five different systems to achieve that. So for us, the whole premise is how do we bring capabilities in one product to be able to serve the needs of the customers? And how do we leverage things like Copilot generative AI or even the AI models, right? And this is the orchestration that we are doing with our product. This is how we want to differentiate. And the customers with whom we’ve been working in recent times, we’ve got amazing feedback both in terms of productivity, in terms of their ability to look at data instead of each time spinning a new dashboard or a new Power BI Query. It is really like you’re working there, you’re doing your planning, you want to ask it some questions, you want to collaborate on creating a consensus plan. You do it then and there. So there is no parallel universe of an Excel and a system with sometimes never meet. So these are things that we are bringing forth with this industry.
Daniel Newman: I really like that. I’ve been thinking a lot about how we all draw our best innovation sometimes by thinking about how we work, and systems and humans have historically not actually worked in a very unified way. If you actually think about it, I wrote a book in 2019 called Human Machine, and one thing I didn’t fully appreciate when I wrote that was the symbiotic relationship that even though machines may not be empathic, if humans can interact with a machine in a way that’s more empathic, that’s when we’re going to start to be able to maximize value. Meaning to what you just said, Sameer, can I talk to the machine in a way that’s natural to me and I can query it in a way that’s like, “Hey, I want to know about this shipment and is it going to make it on time to this location?” In machine language, that’s input code, search query, but in natural language. And we’re seeing this with a lot of the consumer things, but we want our business apps to function this way too.
Sameer Verma: Absolutely.
Daniel Newman: So that’s really encouraging. Now, you’ve been very bold, very ambitious. Microsoft has come out very fast in many cases. When you come out fast, of course your customers are going to be excited. And it sounds like from what you’re telling me they are, but I imagine you’re also learning some things and this is helping shape how you’re going to continue to evolve the product because being first always comes with a lot of questions. So what have you learned from getting to market very quickly with these generative capabilities and really building them across your ERP landscape and how is this going to direct future investments, especially with the Copilot?
Sameer Verma: Yeah, so that’s a brilliant question. So first of all, I want to outline that none of these things that we are doing is behind in the lab. There is a lot of customers who work with us, day in and out. So there’s a lot of trust that companies have placed into us as we evolve this. To give you a sense, when we evolved this entire jobs to be done framework for end-to-end processes, we work with 358 customers, doing a lot of research on how their users are today interacting with the system, what are the points of frustration and all of that. So I think the biggest learning that we’ve had thus far is that there is a strong element of trust. So far, if you look at the machine learning models or AI models, they appeared like black box to users. So they would get a response, and that whole explainability element was completely missing. And I think with Copilot, what has happened is this whole multi-tone, to your point, you’re really talking in a language that you speak and your software is giving back a response, a language that you understand. So for instance, you get in there, you ask a question, you get a response, you say, “Hey, tell me why is this variance happening?” Or “Why is there a delay?” It comes back and gives you that response.
The biggest learning is that you cannot take the most complex problem and put it out there for customers to start accepting. You really need to start with what we call as the low stake skills, and then you move to the high stakes. So as you build the trust, the more and more people start using it, and then you expand the entire horizon of these capabilities. So from our perspective with this approach, what we are really doing is we are talking about this whole AI first mindset, which means that if you are designing any capability or if you’re re-looking at any existing capability, you always need to think about, okay, how’s the interaction of this particular piece of capability with the user going to change? How can we have Copilot go ahead and actually help improve the user’s experience with the software? And this has a very, very different paradigm to it. For example, today if you have a sales order, people do ATP checks, which are complex processes running in the background, but if I’m a user of a software, I can just go and ask, “Hey, can you tell me if I can confirm this order for a delivery on X, Y, Z date?” You simply ask this question and the Copilot gives you the response and you go ahead and confirm the order. So there are numerous such examples. That’s the approach that we are really taking to it, and that’s why we also believe that there is a very strong coming together of solutions into one flow versus sticking for each and every problem a new solution. So how do you bring all of this into a singular flow and leverage the Copilot? So for us, from an investment perspective, it is really about bringing, like I said earlier, the entire portfolio of Microsoft together plus the business process view, and that’s what companies are interested in.
Daniel Newman: So I had a chance to spend a little time with Satya at your New York event a few weeks back when-
Sameer Verma: Yes, I remember that.
Daniel Newman: Announcing Surface and some of the new Copilot features. And I actually asked him, Sameer, when I had just a little bit of time, which was great to have that chance because he’s a very busy guy, I believe, but I asked him about the one Copilot, the one to rule them all. And of course that’s not an overnight sensation that you’re just… But the idea actually to your point, that your apps in your business, at least from a business and your consumer, it makes some sense. Why do we want to be bouncing between ERP and CRM and SCM and HCM and tool after tool and BI, and then of course your productivity? I mean, we need all these things to be orchestrated. We’re going to want to remove all these different layers, layers of abstraction and create a very streamlined opportunity. So great points. I really appreciate that, Sameer. I do want to pivot here though, because one of the areas that the company has started to show some real interest in innovating is one that’s mature, but I think it’s one that I have to imagine your customers were really looking for. You’ve announced some new demand planning capabilities. This seems like an area that maybe Microsoft saw as opportunistic. It seems like an area that maybe customers are clamoring for. What really made you decide to say, “Hey, we’re going to get into this demand planning space and we’re going to go for it.”? And why are your customers going to want to try it?
Sameer Verma: Yeah, so I think that’s a great question, and I can tell you internally we had the same debate as well, right? Because this market is saturated, the long-term players, best of breed, and all of that. But when we started working with customers, we always create a focus group, work with 1,500 customers, try to understand the problems, et cetera. One of the things that we realized is that these algorithms have existed forever. A lot of companies have invested in data scientists, their own machine learning models, et cetera. But on the other hand, there’s a parallel universe that exists in Excels where a lot of changes happen in theses Excels. There are hundreds of versions of these Excels where people do their forecasting, their consensus forecasting and then finally approve this. Many a times these two worlds don’t meet. And the next thing is the whole preparation for any kind of strategic planning activity or a sales and operations planning, it takes months. So what we did really was we said, okay, our mantra is it’s going to be simple, it’s going to be fast, it’s going to be flexible. So through that approach, what we’ve done, there are a couple of really differentiating points. So first of all, we are not really trying to separate the supply chain planning and execution world. That’s why it’s a very, very deliberate move to bring this capability inside our existing Dynamics product. It’s not a separate suite. Customers are going to get it where they are. The second thing is we have leveraged a lot of AI models, but at the same time, we are also allowing customers to bring their own models with, because customers have spent sometimes millions perfecting those forecasting models and they don’t want to drop it. So we allow them to bring that.
The other aspect of this is that collaborative planning, the reason people have been using Excels is that they put an Excel out there, they use SharePoint, they collaborate on it, but then they fail to bring all of that back into the system. For us, what we are doing really is in the system, you have your Excel, you collaborate, you make changes, there’s complete auditability, traceability of all the changes that you’re making. People can comment on it, agree at the cell level, and then boom, you push it back. It’s there in the system. The other aspect is also that, like I said earlier, with this demand planning, we are really bringing a strong collaborative element. You get your forecast, you ask it questions, it tells you why something is this way or that way. So it is really bringing together all elements into the solution and then a very seamless plugin with your entire supply chain execution. And that is why we believe we are going to differentiate a lot and the signals that we have from the customers, some really, really complex customers that we’ve been working with, they’ve said it’s like mind-blowing. For them, the amount of time it has taken to do this whole prep work to what they can do with the solution today. That is what makes us super excited and convicted about making this very successful.
Daniel Newman: So let me double click on that. We only have a few minutes, but when I first saw what you were doing with demand planning, what I immediately liked was, wow, this is a really high power, lightweight solution for companies that probably don’t have much of a demand planning capability. And boom, they can plug it in and add it. But as I spent more time, what I really was encouraged by was, this is actually a really powerful tool that can work side by side with the investments that companies have been making in demand planning. Talk a little bit about that because I’m concerned as an analyst, one of the things I was concerned about is people are going to mistake this as just a lightweight sort of plugin, and they’re not going to see, oh wow, this is something that actually we can implement side by side with our large investments in demand planning software, FP&A software, talk about that and how best of breed systems today and what you’re offering are being brought together because I think that could be a really big opportunity for Microsoft and for your customers.
Sameer Verma: So that’s a great question, Daniel. And we also realize that it’s a starting point. So we are not at all prescribing to customers that they need to immediately rip and replace everything that they have. So it is also a lot about taking data from these existing systems, bringing it in our system and going through the entire process. That’s number one. The second thing is that there is an evolution. Today, we are bringing certain capabilities into the market, which are really focused on the demand planning aspect, but companies also need to do, since it’s an operations planning, et cetera. Now what we are doing, for example, as the product evolves, our hope is that customers are gradually going to move workloads till then these things need to coexist. There’s also a secondary aspect. Our focus is very horizontal in nature. We serve manufacturing retail, CPG professional services as our core industries, but there are certain best of breed players out there who are very specific on very industry specific capabilities. So for us, this is also an opportunity to let these co-exist as far as they can. But from a broader ambition and aspiration perspective, we are going to be launching our xP&A capability in December. So this is when your entire demand planning, your financial planning, all of these come together in the realm of a product. Now, we are super aware that no customer is going to be just pure a hundred percent Dynamics, especially if you look at larger customers, federated landscapes. So for us, the idea is to allow them the ability to bring all these things together into one system and then give them analytics insights. So it is not really an either or scenario. For some category of customers, yes, it makes sense to pivot to one, but for many it could be that these coexist and we are fully aware of it. That’s why we keep it very open. We have a platform approach. You can plug and play, you can bring data from other systems. So this is actually the beauty of the approach that we are outlining there.
Daniel Newman: Sameer, I think that’s a great way to wrap it up. A future where systems are connected, collaborative, capable, and natural in terms of how we interact with them. And that’s very encouraging. That’s very exciting. And I just want to take a moment and say congratulations on all the progress you’re making. Of course, I will be watching closely and providing my perspective as you continue to launch new innovation and in a competitive market, that’s what we do here. We look at the market. But what I’m seeing so far is very encouraging, it’s showing that, A, you can really build an ecosystem around what Dynamics does. But you said it really well at the end there. Many companies have many systems and building what’s augmenting technology that uses things like Copilot, natural language. That’s really where the future is going. So congrats. Let’s connect again soon. Let’s talk more about the innovations that come next. But for this one, thanks for joining me here on the Futurum Tech podcast.
Sameer Verma: Thank you, Daniel, and we look forward to your critical feedback as always.
Daniel Newman: All right, everybody hit that subscribe button. Join us here on the Futurum Tech podcast for all of our interview series and all the great conversations we have with executives from across the tech landscape. For this episode, it’s time to say goodbye. We’ll see you all later.
Author Information
Daniel is the CEO of The Futurum Group. Living his life at the intersection of people and technology, Daniel works with the world’s largest technology brands exploring Digital Transformation and how it is influencing the enterprise.
From the leading edge of AI to global technology policy, Daniel makes the connections between business, people and tech that are required for companies to benefit most from their technology investments. Daniel is a top 5 globally ranked industry analyst and his ideas are regularly cited or shared in television appearances by CNBC, Bloomberg, Wall Street Journal and hundreds of other sites around the world.
A 7x Best-Selling Author including his most recent book “Human/Machine.” Daniel is also a Forbes and MarketWatch (Dow Jones) contributor.
An MBA and Former Graduate Adjunct Faculty, Daniel is an Austin Texas transplant after 40 years in Chicago. His speaking takes him around the world each year as he shares his vision of the role technology will play in our future.
