Name: Unlocking Working Capital With AI-Driven Procurement
Uploaded: 2026-10-20T16:16:11.739Z
Duration: 1 h 8 s
Description: Unlocking Working Capital With AI-Driven Procurement

Transcript for "Unlocking Working Capital With AI-Driven Procurement": Before Amazon Business, buying for work was chaotic. Now it's easy to find products from thousands of suppliers in one place. Save on every type of purchase from individual items to bulk orders. At any time, you can view your spending on pre built easy to use dashboards. Plus, you can free up cash flow if you choose to extend payment deadlines and view and approve your team's purchases easily. With Amazon Business, things just got a lot more manageable. Today's webinar is presented by finance chief in association with Amazon Business. Our topic is unlocking working capital with AI driven procurement, improving cash flow, visibility, and financial agility across the enterprise. With economic uncertainty and tighter margins, working capital has become a strategic priority. Procurement decisions directly affect cash flow, yet procurement and finance teams often lack shared visibility. AI is starting to close that gap. It can analyze spend in real time, improve forecasting, optimize payment terms, and free up trapped capital, turning procurement from a cost cutting function into a strategic lever for liquidity. Today, we'll look at how AI is changing procurement, where organizations are seeing measurable value, and what leaders should prioritize next. Joining me is a specialist in procurement transformation, financial operations, and enterprise strategies. Please welcome Frank Sirica, senior partner, business operations practice leader at IBM. Hi, Frank. Thanks so much for being part of today's conversation. Yeah. Thanks for having me. I appreciate the opportunity. Our pleasure. So I've given you a a little bit of an intro there, but I was wondering just for the people at home if you could maybe just introduce yourself introduce yourself, a a bit more... In a bit more detail, please. Yeah. Sure. So, Frank Sarica. I'm based in Research Triangle Park, North Carolina. I'm a lifelong procurement transformation leader. I I think I've been doing this since the beginning of my career, kinda started as a practitioner, as a buyer, moved my way up into strategic sourcing, went and ran a delivery center off shore for a couple years. And then, really, the last, you know, ten years has really been focused on helping clients, both in finance and procurement across source to pay really drive transformation. And it's an exciting time right now with, you know, all these things, everything AI and how it works changing. And, it's it's been great to kinda see our clients transform as well as the technologies that help enable that. You're absolutely right. It's a really exciting time. I mean, one of the major shifts many organizations are experiencing right now is that working capital optimization is is no longer just a financial issue. It's becoming a a broader operational priority that's increasingly involving kind of your procurement team, your supply chain team, and even your executive leadership teams kind of working more closely together. So I was wondering just to kick us off, what factors are driving greater focus on working capital management today? Yeah. I mean, I think how we look at this is, like, we've been through a series of, like, overlapping shocks in this space, and each one of them exposed to different vulnerability and how organizations are really managing capital. So, I mean, the the big elephant in the room is really with inflation and interest rates. I mean, when the cost of capital rises, every dollar tied up in excess inventory or slow moving payables or unoptimized payment terms has a real and measurable cost. So CFOs who once kind of tolerated this, because money was cheap are now much more focused on it. So when we think of that as, I would say, a big foundational, highlight in the area. And then layered on top of that, we've... We still see that supply chain volatility. I mean, the disruptions of the last several years, have really caused companies to carry more stock and inventory as a hedge, but that ties up an enormous amount of capital. And so, like, we work with our IBM's Institute of Business Value Research, and it consistently shows that supply chain resiliency and financial agility are really two inseparable items and priorities for CFOs and CEOs alike. And then, of course, you know, we get into the way companies are working today. I mean, the margin pressure story is amplified everywhere. I think everyone's looking at how to drive top line growth, and that's become harder to sustain with things like inflation and cost of cash. So, you know, shifting on that attention on operational efficiency where procurement and working capital kinda come together, it's really become a source of competitive advantage, not just a cost center item. Absolutely. And do you see... You you kind of outlined there this... The shocks that we're dealing with kind of daily, you know, once in a lifetime crisis seems to be a kind of monthly occurrence now for procurement supply chain teams. So is that... Do you think that's kind of where, like, maybe legacy procurement models are now creating an efficiency that having an impact on working capital performance? Yeah. I mean, I think data as a whole has been a barrier. And when we think about legacy systems, you know, we have fragmented spend data because you have these, you know, most or large organizations have multiple ERPs, business units, geographies, and there's not necessarily unified view of what's being spent, who's... With who they're spending it with, the terms can sometimes conflict across those. So it becomes really challenging to make those intelligent decisions about payment optimization, contract leverage, and even supplier consolidation. I think we are moving closer to some optimization here, but it's still been a big pain point. The other areas, procurement manual, most of it is still manual and somewhat reactive. I mean, I don't think we all have, you know, demand forecasts the way we would like to see. So we lose that ability to optimize order timing, quantities, even, like, if you're thinking about freight. It all becomes a challenge, and emergency purchases, you know, typically come with worse prices and less favorable terms. So we... Those are all things that we're seeing that are impacted by these legacy procurement models across the organization. So is it fine... You mentioned the the kind of your your foundation, your data, the kind of, the forecast models you had. Are those the reasons why organizations are often struggling to kind of gain that kind of accurate real time insight into the procurement related financial exposures or there are other factors that play kinda leading to that? I mean, I do think there's other factors, but I I do think the core problem is really around data architecture. I mean, this is, you know, compounded by organizational alignment. So where we're seeing more optimization today is where we're seeing more connected, like, in procurement. It's not just procurement. It's an integrated source to pay process. And we wanna make sure where that exposure lives in those multiple systems, we're getting the right people in the organization. So I've seen where I've talked to a procurement organization. They've run, like, optimized payment term projects and moved everyone from, you know, thirty to sixty to ninety day payment terms, sometimes one twenty. But then what's happening in the back end of that system, and how are we doing that? But then there's another factor now that we should always be looking at is, like, well, what are the what are the upsides for maybe paying quicker, and are there discounts that could be achieved. So those pain points kinda live across the organization, but unless you you are really talking across the organization, we have a thing at IBM. We kinda talk about all of our data being on the glass. Really meaning, like, we're... It's one view of the truth. No one's coming in with their own report. They're looking at things collectively, and those decisions are getting made based on that one source of truth. You know, so whether you're talking about, you know, your financial close, your quality forecast, all those things are consistent across the organization. And that's a big... That was a big transformation for us, but it became foundational in the way we work as an organization. So, Frank, yes. So how is AI improving procurement visibility and helping teams to identify working capital opportunities? Yeah. So I think the biggest thing is really thinking about how we take both structured and unstructured data from everything, from invoices to contracts to to other payment terms that we have in the organization. So anywhere where we can get that and have that unified view, that really creates that that that opportunity for not only patterns, but where we could basically shift to include... Improve that cash flow. So, you know, one of the areas is if we have tens and thousands of suppliers and millions of transactions coming across multiple systems, how do you harness that in a way that can be reactive? And that's really putting, you know, not only working capital, but better buying and supplier management on a better spotlight than what we had in the past. And we've seen that data capture and putting that in actionable insights for our clients and ourselves to be the the key component of how we're driving change in the organization. So when you're able to kind of optimize your your data flow, how can you then, you know, use that alongside AI to kind of help improve your forecasting accuracy and make more... Become more proactive than than maybe reactive when you're making your kind of financial decisions? Yeah. So, I mean, I think forecasting is the area where we have... Especially with using AI has the most transformative impact on our finance procurement interface. So because of the accurate forecasting is in the foundation of everything else, like, when we think about cash flow forecasting, it's been limited by really the quality of the input data and the modeling. And now, you know, and then you shift to the finance teams that have typically relied on the procurement data, whether it's complete or incomplete or sometimes delayed. It requires a lot of manual interpretation. And most of our organizations have struggled to really see and capture the complexity that exists in our modern supply chains. So instead of discovering a cash flow gap at the end of the month and scrambling, we're seeing that typically weeks or months in advance, and we're able to take those deliberate actions, whether that's with a supplier, whether that's a a putting in a discount program, negotiating some terms, or potentially rebuilding some, inventory. So that's how we've really seen those accuracy improvements across the organization, And we're seeing improvements in the range of 15 to 25% with AI enhanced planning, but I think that's a number that can essentially grow. And if you also bring in other demand signals like your sales, pipeline, which is kind of the next wave of this, I think we see it, more optimization in this space. So you spoke a lot there about kind of the forecast accuracy. What other kind of procurement workflows are delivering the greatest efficiency gains through, you know, the use of AI and automation? Yeah. I mean, I think the the area we started first and maybe it's because it was easier is really accounts payable automation. I mean, this has... We think about the velocity of this work and, you know, there was a time where we'd still have to go chase blocked invoices, kinda understand what didn't match, etcetera. And now we're using AI really fully autonomous to drive those those exception workflows across the organization. So I would say, you know, using our intelligent automation capability to reduce invoice processing, we've kinda seen upwards to 70 to 80% reduction in that audit... With that automation. Also, like, on the front end on the purchase order processing and approval workflows, it's another high value area, very similar but more upstream. You know, being able to look at routine approvals based on policy rules, flag exceptions for reviews, surface information about, like, supplier risk, budget status, all those things. Those were a lot of manual checks in the past that now are much more automated. And then the last piece really being contract life cycle management. I mean, we talked about being extract, able to extract metadata from our contracts and now linking contracts to payments, being able to make sure we're aligned on those payment terms and those dollars really give us an opportunity to eliminate that value leakage across contracts, which we typically see can be anywhere from one to 2%, which is quite significant if you think about total spend. We've spoken a lot now about the kind of the data, the numbers, how it can improve the kind of financial terms. What about how AI can impact to make your supply management more more effective? How is it impacting on those more kind of personal relationships that you have with those kind of key suppliers that help run your business? Yeah. So I think the supplier management and payment strategies are basically, like, two sides of the same coin from a working capital perspective. So I think AI is helping organizations think about them more holistically. So on the payment strategy side, you know, AI brings real sophistication of... To what's been kind of, like, a blunt instrument of managing this. So, you know, now we apply... You really can think about applying the right terms, whether it's that 30, 60, 90. And AI allows you to make a more granular dynamic approach and really segment your suppliers to get the most impact. And, I mean, you can chase the work kind of becomes the same for a lot of different suppliers. So how do you do that to where it makes the biggest impact on your working capital? And then also, same thing if you look at that on dynamic discounts, like, where would you get the biggest bang for your buck from doing that? So I think those programs can typically deliver, you know, now that cash is so valuable, it can impact eight, 15% of your deployed cash, which is great. So as you... We've spoken a lot about AI and how great it can be. So a lot of people thinking maybe we'll just go install some AI now, put it into our workflows, and and we'll be away. Obviously, we both know it's not as not as simple as as plug and play with AI. So I I wonder if we could look at, you know, when you're looking at what kind of slows the process down when organizations, you know, maybe attempt to modernize procurement of finance operations through through the implementation of AI? Yeah. I mean, I think data readiness is always a start... Your starting point. So... And data will never be perfect. So I think we have to all move past that. But, you know, how do you get your data harm... Harmonized, get it workable, and then really work on what maybe could be enriched in that dataset. But if you have a clear view and you're looking at, again, that one source of the truth, I think that really becomes your foundation across those and just making sure that those systems are talking with the AI. The other piece, of course, we talked about in the past very siloed organizations. So how do you bring that organizational alignment? So procurement and finance transformation requires this kind of collaboration, and it... It's... Can be difficult to sustain. So we have competing priorities. We have different reporting lines. We have different concerns about the the automation and AI and the impact of the roles within the teams. But you really have to work as a change function to drive that capability building across those dimensions because the technology investment without the organizational change will not maximize the potential of this as a whole. I I mean, I'm so glad you mentioned silos there because, you know, from what I've seen, the organizations who have the greatest in... Are delivering the greatest value from procurement transformation are often the ones treating procurement finance and and operational planning as part of a much more connected ecosystem, you know, that requires not only technology investment, but also stronger alignment around data, governance, and decision making. So how important is alignment between procurement, finance, and supply chain teams when improving working capital performance? I I think it's extremely important. It's foundational because technology is really the secondary factor we should be looking at. So the organizations, you know, that are see... Seeing the greatest working capital improvements are the ones that have established that alignment between procurement, finance, supply chain, or at least in parallel with the technology work that's happening. So the alignment being shared KPIs, shared, shared, visibility across those metrics of working capital. So it shouldn't be just like savings percentages that we typically see in procurement. It should really be about what are the outcomes that are delivered across the organization. And we at IBM have this, you know, philosophy around savings and procurement. It's like an IBM save is an IBM save. It doesn't really matter who's counting it. And that's the way we're kinda shifting that that mindset. So when we think about transforming those organizations, I think you have to look at, you know, your operating model and do, you know, be honest with yourselves around how it will work because if if you're not working alongside the humans within those teams, the technology is an accelerant, but it doesn't necessarily drive the optimal, outcomes without the alignment with the teams that are that are... They're supporting. Absolutely. And once you've have that AI and you're adopting it and we're seeing more and more companies bring it on board and and bring it into their workflows, So what do you believe kind of governance and oversight considerations should be made a priority for organizations as they as they implement AI more and more into their workflows? Yeah. I mean, governance has to be foundational. I mean, it, to me, it's it's where a lot of problems can really kind of rise up very quickly. And we've seen this with, you know, a lot of teams create bots, and now you have bot sprawl kind of in the organization. And what happens if one breaks down and it's not supported? I mean, we've seen those problems. So with the governance piece, I think AI, it does take on much more consequential sequential roles within the organization. So if it's recommending payment strategies or flagging supplier risks, I mean, we should really know how that... Those decisions are being made and making sure that we're very well informed on, you know, govern... Governing AI just like any other business process. So define decisions that AI can make autonomously, align on decisions that require some amount of human confirmation, and those requiring even a more senior review. And build those controls into the AI workflow from the start because it's harder to come back after the fact to retrofit that governance when when the, AI is already built out. Do you find that kind of oversee... Governance is maybe the biggest roadblock of implementing AI, or is there other things that procurement leaders particularly kind of find themselves stowing against on this kind of road towards a kind of more autonomous, procurement life? Yeah. I think it's more of a... I think you have to look at things holistically. And I think part of the problem is everyone's challenge to deploy AI and do this quickly because we're all dealing with some of the OPEX challenges that were in the team. But I think we really need to make sure that we're doing that while maintaining the... Not just the the optimization of of the process and the and the teams, but also how you do really drive that capability and make sure the teams understand how to work with AI collectively. I feel like we're hitting a point to where the AI readiness is is is a challenge more in the process teams that manage it today, and how do we bring those two together to have a more AI led, skill set within our teams to really manage and work alongside AI or digital workers as a whole. And I I think that is becoming one of the biggest... Bigger inhibitors that I'm seeing in the last few months. As an organization kind of undergoes this modernization of their procurement operation, how can they maintain agility and operational control while some have taken such a kind of a massive transformation? Yeah. I think it's... You have to do this in a phased approach. I mean, think about it as, again, value led rather than a wholesale platform replacement. So, you know, if we think about modernizing, following a similar pattern and identify two to three high impact, well scoped AI opportunities, implement those quickly to generate a visible ROI, and then build on that success and the organizational confidence and and use those dollars to invest for broader transformation. Those early wins really become the key to maintaining momentum. We see it... We hear it all the time. We have 20 AI pilots. Well, how many of those pilots never make it to production and why is that? And I think it's often because we haven't really thought about those opportunities collectively and how they impact the organization. So it's not just the technology, it's really being... Thinking about that from an overall implementation and and roadmap across the operation as a whole. Progress, I wonder if you could maybe gather your your your crystal ball for me. I know it's kind of hard to make predictions as this transformation is is undergoing kind of minute by minute right in front of us. But, you know, what changes as as we sit here today do you expect to see over the next maybe three to five years in how procurement supports financial strategy? Yeah. I think we're gonna see those lines continue to blur across finance and procurement. And the data and technology infrastructure is maturing to the point where it's more real time, we have shared visibility, and there's much closer integration of those teams. And often because we're both sitting under CFOs in many organizations today. So I think that we we see more joint activities versus handoffs that we've typically seen siloed in the past. The other piece is, of course, you know, moving beyond AI agents to more autonomous roles. So truly agentic AI where AI systems plan and execute multi step workflows without constant human intervention. This is already being applied to many of the procurement context, but I'm expecting this to really move more upstream, whether that's handling routine supplier task, negotiation tasks, looking again, we talked about the dynamic payment optimization, demand driven procurement, all those things, humans being involved but becoming less of a more exception based than, a true part of the process. For those organizations perhaps earlier on in their their transformation process, where should they focus first? Yeah. I mean, I think, again, getting back to data, doesn't need to be perfect to start, but you need to understand what you have, where it lives, and what are the highest priority integration gaps. You know, an honest data assessment, even if you do a lightweight one, will save you a lot of pain downstream. Next, I'd... I'm back to those one to two high impact well scoped use cases. Again, I think AP automation is the right starting point for many. It's it's a pretty defined process already. The technology is there. The ROI is measurable. So it it seems like an early win, to build some organizational support for broader transformation. And then the third thing is, again, build on those cross functional teams. Establish that joint governance between procurement and finance, even if it's informal at first. But talk about what the technology investments you're both interested in, how those could work better collectively across the organizations, and and then help prepare your organization for that readiness. Not just technology readiness, but that organizational alignment that has to happen across those teams. How do I go about addressing maybe AI hesitation within my team? There's maybe some people who are on board with it, want to begin implementing it. Others are maybe a bit more hesitant of of introducing this new technology into into our workflow, sharing our data with it. How do we go about kind of easing that that that hesitation? I think that's, I think that's more of how you communicate around how AI can really help them be more impactful in their job. I think I think a lot of people do have fear because it is changing and we're hearing about it in so many facets of our lives that it can be scary. But I think helping them work with it, understand how it works, what some of the limitations are, and offer more of the upskilling. Because I think there are people who just feel like they need to be highly technical to work with AI, and that's becoming less and less the case. We have people who are not technical at all. I mean, I can create an agent now using some of our tools that we have that, you know, just by prompting. And so I think the more people use it in their day to day life, even if it's in, you know, again, personal life, but also bringing it into bite sized bits into the way we work, opens up some opportunities for people to look at it differently. And instead of looking at it as a... As someone who's going to replace the work I do, but someone who's going to make my job easier and allow me to make a bigger impact for the business and the stakeholders I support should be the mindset that we all have. And how do we ensure that that first roadblock doesn't send the whole, journey off course and and maybe stop the pilot or stop the ambitions reaching the heights that they can do? Well, it's a... I'm not really sure how... You know, you will always have someone who is not for it and who doesn't wanna work in that way. And I think you always have to assess your workforce in in that manner. But I think for the most part, when people understand where it's going, how it's going to be used in our part of those teams, so I think, again, that collective communication and try to make it something that's not happening to them, but they're part of it. I think that really breaks down those barriers for teams. Thank you, Frank. And I think we'll all agree that maybe I've asked one too many questions now. So I have a few from the audience ready to go. So I think I could put a few of them to you, Frank. So the first one I've got is, which AI capabilities are generating the fastest financial impact in procurement today? I mean, I think, again, I'll go back to the accounts payable automation. It's quick. It's fast. It has a large people impact, and it's typically not been something that I think people necessarily want to do. So I think we're finding a much broader opportunity to reskill a lot of the the teams that have been doing that work in the past, and they're moving on to more upstream work. I think as we move right behind that, it's really AI driven spend analytics. So being able to see those patterns, I mean, that is more upstream. It's driving much more strategic value in the organization, becoming more in... Moving away from something being more backward looking to more forward looking. And again, whether you're talking about working capital where you can identify work, you know, things like early payment opportunities or, some of the negotiation of payment terms, but also just how you're doing supplier consolidation, improving some of your spend and visibility with some of your key suppliers. I think that has become highly valuable, in the organization. So I kinda like one that's coming, like, more downstream and then, the spend analytics more upstream in the process. Thank you, Frank. Another question we've got is how can organizations improve visibility into procurement related working capital? Yeah. I think the first part is the integration. So connecting those teams into a single real time model instead of leaving them in the separate systems, I think, is key. Once you have that foundation, AI can continuously analyze the data and surface working capital insights that rather than requiring those manual reporting cycles. So our approach at IBM has been, you know, combined with an intelligent data layer with AI analytics over top of it. It gives both the procurement and finance team a live view of those spend commitments, payment obligations, working capital exposure, and organizations that we... Like us have done that, we stopped treating working capital visibility as a report, but really as a continuous operational capability in our organization. Perfect. And our final audience question is, how can businesses improve cash flow without negatively impacting supplier relationships? Yeah. I mean, that's a great question because the instinctive approach is simply extending payment terms to preserve cash, and it does. We've seen it backfire by damaging supplier relationships or increasing supply risk. So a better path is always a value sharing model. So dynamic discounting allows suppliers who need early payment to access in exchange for a discount. It's voluntary. It meets the supplier's financial need, and it delivers a return for the buyer. Right? We also think, like, when we think about our supply chain finance support in this approach where a payment strategy is optimized for both the buyer and working capital position and the supplier's financial health. So if it's done well, it actually can strengthen the supplier relationship because suppliers see you as more of a partner, not just a customer extending payment terms. And this relationship quality pays dividends when you think about priority, flexibility, resilience, and especially when supply markets tighten. Perfect. Thank you, Frank. Unfortunately, that's all we have time for today. A sincere thank you to Frank for sharing his expertise and perspectives. One of the central themes from today's conversation is that procurement is becoming increasingly important as a driver of financial agility and enterprise procurement. As organizations continue operating in more volatile and cost sensitive environments, stronger visibility, predictive insights, and closer alignments between procurement and finance will become essential. AI is helping organizations move beyond reactive procurement models towards more intelligent and financially connected operations, improving cash flow visibility, supporting smarter decision making, and unlocking greater working capital efficiency across the business. Thank you again for joining us, and we hope you found today's session valuable.