AI-Powered Vendor Management and How CFOs Streamline Operations

Megan - 00:51: Today, my guest is Sara Wyman. Sara is founder and CEO at StackPack. StackPack is an AI-powered platform that fully automates and manages your team's software, services, and contractors starting at just $99 a month. Their mission is to connect the world's companies with the right partners at the right time on the right terms. Prior to founding StackPack, Sara was VP of Strategic Partnerships at Affirm Inc, where prior to that, she held positions as VP of Enterprise Sales and Senior Director, Head of New Market and Business Development.

Megan - 01:24: Sara, thank you so much for being my guest on today's episode of CFO Weekly.

Sara - 01:29: Thanks for having me.

Megan - 01:30: I'm excited about this discussion today. We're going to be talking about the CFO's role in vendor management and taking a look at AI use cases in contract term extraction, forecasting, and risk alerts. I'm looking forward to learning about you and from your experience, so let's get started. First, before we dive into AI and vendor management, can you share a bit about the inspiration behind starting StackPack, and what was the moment that made you realize that this was the right business to launch?

Sara - 02:00: For your listeners, we're building an all-in-one solution that helps companies manage their vendors. So this is from vendor discovery and evaluation to gathering internal approvals down to managing renewals, spend, and contracts. Honestly, the idea came from my time kind of living and breathing at Etsy and Affirm. These are high-growth companies making their way toward IPO. We were building so quickly that we just began partnering with hundreds and then thousands of vendors, and we didn't have time. We didn't manage the spend. We didn't track status. We didn't really assess risk. And then it got to a point where it really became chaos. We got to a point where there were many, many, many more vendors than there were employees, and we see that across the board nowadays. It was amazing to me that there was no system of record. Every company I found has a system of record for their people. You have Workday. You have Rippling. You have a system of record for your customers. You have a CRM, Salesforce, HubSpot, whatever it is. You even have a system of record for your spend, QuickBooks, and NetSuite, but really nothing for vendors—the companies that are powering your company. And that was just amazing to me. Looking back, I think it was so manual. But with the emergence of AI, we don't have to do that anymore. AI can do a much better job, a much faster job at most of this work. It really was a hypothesis that AI could solve the problem, but it became a moment. Really, as I began to interview companies of all sizes, of all verticals, very few of them had a solution in place. If they did, it was a solution that certainly wasn't automated. Honestly, most of the time, it was a spreadsheet. And so we really found that every single company in the world has this problem. And so as we started to rack up customers before we even had the product in hand, that was the moment of, like, wow. We need to go build this thing. So, yeah, we started with vendor and contract management. And today, StackPack helps control costs. It saves time by automating vendor spend, contracts, and renewals, all on behalf of finance teams. And, honestly, it only takes a few minutes to set up.

Megan - 04:08: And who is the ideal client for this product?

Sara - 04:12: We mostly work with high-growth companies. So a lot of them are VC or PE-backed, Series A through D, and we do primarily work with the office of the CFO. So CFO, VP of Finance, Accounting Manager. We find that vendor management starts and ends with finance. And then, typically, what we will see is IT managers, legal teams will obviously get involved in that process.

Megan - 04:35: As a CFO, what role do you believe financial leaders should play in vendor management, especially when it comes to ensuring cost efficiency, risk mitigation, and strategic alignment with the company's long-term goals?

Sara - 04:48: I think the role of CFO has changed so much over the last few years. We've moved so far beyond traditional financial oversight, and I actually really feel for CFOs. The expectation is that they are true strategic partners to the business, to R&D, to the CEO. And so their job, obviously, is to manage costs, but it really is to see around corners now. Because we're seeing all these modern companies with hundreds of vendors and third parties, I mean, financial leaders have to be involved in vendor management in a proactive way, and they can't ignore it. They have to spend time on this. And so, obviously, that starts with cost efficiency, but it touches risk mitigation and strategic alignment with company goals. And, obviously, the big thing that everyone's focused on right now with all these AI tools is time savings of its own team to be able to maximize output. So as I said, I think vendor management starts and ends with finance. And just to really paint the picture of where they get involved, at the very beginning, they set the budget and obviously track it. And then finance will evaluate, and they'll approve vendors, obviously, for cost, but also for company fit. And more than ever, it's risk and it's vulnerability mitigation. They work really closely with IT and legal, and it's not siloed anymore. I think it used to be. It's not anymore. It's very collaborative. Finance then helps negotiate favorable terms. They care about our vendors' seats being utilized or their ability to cut costs throughout the year, and they obviously want to ensure that renewals aren't missed. So, ultimately, I think CFOs are calling the shots on the entire vendor management life cycle. And that life cycle has gotten a lot more complex. Right? As you're adding in all this new tooling, you're adding in data privacy issues. You're adding in brand-new pricing structures with these AI tools that we've never seen before. A lot of utilization-based pricing. So vendor management, it's like becoming its own category for CFOs. It really is. It's much more strategic than simply cost-cutting now.

Megan - 06:50: I'm just curious. When did you launch StackPack?

Sara - 06:53: We started building StackPack in August 2023. We came out more publicly in November 2024. So we've been out publicly for the last four months.

Megan - 07:05: Well, I'm sure the need's only grown with the staffing shortage and, you know, people just finding talent wherever they can these days.

Sara - 07:13: When we launched this, it felt like CFOs were careful and maybe a little nervous around the use of AI. And, yes, to your point, that has changed completely in the last year. I think people are very excited to see how they can leverage it in this staffing and in this resourcing environment.

Megan - 07:32: And how do you see AI transforming vendor management, particularly when it comes to automating the extraction of contract terms and managing compliance effectively?

Sara - 07:42: I think as mentioned, AI can take a lot of manual work off teams' plates. You mentioned term extraction. I think that's a brilliant use of AI because it's faster. It's more precise than human eyes are. We've personally trained up our AI to be able to review MSAs and order forms and contractor agreements, invoices, whatever it is. And we can pull out start and end date, renewal dates. We even do the math on the notice dates. We can also extract, like, "Hey, what is this exact product and service that I'm purchasing? What's the product name? What are the cost terms? What are the units? Is it seats? Is it licenses? What is it?" We can also pull out if it's variable or fixed pricing. So that's on the cost side. We also can extract legal terms, like I mentioned, data privacy, security, things that companies care a lot about. And so by extracting all those terms, the very first thing you obviously do is save time. So we hear from teams that if they were to manually do that, that takes hours and hours of work. It requires additional headcount. So a lot of times, they don't just do it, and they obviously pay the price, right, on missing renewals or they get bad terms for the business. So the first thing you get is time savings. The second thing it does is allowing for an automated, precise system of record. I think you mentioned compliance. Compliance officers love this because the output from the extracted terms, they end up knowing every single third party in the system. They have eyes and ears around them. Everything's in one place, and they really have a sidekick, I think is what we call it, to help them do their jobs and assess risk and compliance. I think the last thing is that once all those terms are extracted, you just have all this data, and it's pretty endless what you can do with it. You could start making recommendations. You can start actioning on them. You could start comparing your terms to other companies' terms that look like you. You can analyze them. So where we started, very simply, was on alerts. So we alert for a new vendor that hits the system. We alert for a really big, meaningful increase in spend that maybe caught us off guard. And then we also alert on notice dates and renewal dates. What I love about it is AI can triage all of this for CFOs and make sure they don't miss everything by, obviously, raising an email or Slack or whatever the mode of communication is there. But, yeah, we see a world in which, for compliance, AI can be used to just alert on vendors being out of policy or maybe there's some risks on a number of vectors that the compliance person cares about. So just in our opinion, endless use cases here.

Megan - 10:18: I'm excited about AI. I use ChatGPT, and that's about the extent of my AI. But it's amazing. So how can AI be leveraged in financial planning to improve the forecasting process, especially when predicting costs associated with vendors or other external partners? And what key metrics should CFOs focus on?

Sara - 10:40: So we use AI in a few ways to help CFOs forecast. We focus on software, services, and contractors, so really any third party. I think there's a number of steps you can take. It's just one, aggregate a list. Create that system of record that we talked about. We use a variety of sources like accounting systems, contracts, our own proprietary sources. You sign up with us, five minutes later, you have a magical list. Honestly, that's a great first place to start, a system of record. Second is then taking that list and applying the actual spend to it. Look at the historicals. It's a good starting ground to see where you are. And then the third step I'd say is really where it gets interesting is using AI, taking all those extracted terms, and then you can start forecasting. And you can start forecasting using actual contracted spend. This is spend that you must pay. You're legally obligated to pay it. It's super, super helpful as you think about forecasting and use of cash. So a really interesting thing there. And then I'd say the fourth thing that we're really excited about is using AI to help forecast that variable spend. So where spend will increase or decrease with seats or maybe payment volume or maybe total headcount, you can get very, very precise because all your data's in one place, and it's very easy to manipulate. So, again, I think you'll find AI is going to be really interesting and invaluable in this domain. The other thing I could just call out is in the past, you would create your forecast or maybe it's your budget review, and it's all in an Excel doc. You use it. The next day, it's stagnant and it's out of date. I think with AI, it's always going to be updating for you. So it's constantly real-time. It changes as your business changes, so the accuracy increases. So that's what excites me there on the forecasting side. I think you asked about key metrics. We monitor three things. I'm sure there's way more, but just three I'll call out here that we really like to track when it comes to vendor management. One is cost per employee so that you can really look at the unit economics of who is using this tool and how to think about variable expense. So that's the first one. The second one is ROI of the tool of service. Basically, the cost of the tool versus the intended output. It's pretty hard to do. It's pretty manual right now, but I think AI is going to make that easier as we think about, "Did this tool make us faster? Did it make us smarter?" I think we're going to be able to tie that back to the actual cost. And the third thing that we do track is just criticality to the business, and that's obviously not a formula. But we're able to tag vendors so that everyone in the company knows if they're strategic and how do we treat them. So a few ideas there. I'm sure there's a lot more, but we're, again, pretty bullish on how AI can help, obviously, with vendor management, but also with forecasting and analytics.

Megan - 13:31: And I'm curious, but once a customer or client decides that they need this tool, how hard is it to start using it? Like, what is the process for getting up and running?

Sara - 13:42: So we like to say that our process only takes five minutes to get on board. This is true, right? So a customer will log in to our system. They will connect in their accounting software, so QuickBooks, NetSuite, Sage, whatever it is that you're using. From that, we get all your vendors, and that's when our AI jumps in and starts basically filtering so that you can see the vendors that actually are managed, right? Software, services, contractors. And then you're set up to go. And then we do help you get contracts into the system as well through a wizard that we have, but it truly is a five-minute setup.

Megan - 14:14: And you mentioned that AI tools can provide real-time alerts for potential risks. So how can CFOs ensure that these risk alerts are being used proactively rather than something that they're just reacting to?

Sara - 14:27: AI does a really good job of the alerts. Like you said, I think if it falls on deaf ears, it's not that valuable. So, like, an example that we found was that when we alert customers on new vendors that hit the system, maybe that's coming through on an employee credit card or accounts payable. CFOs like it, but it's not overly actionable. It's more of a reaction of, "Oh, gosh, here's another one. What is this?" Right? It's kind of noisy. It's a lot more valuable when there's an immediate call to action or it's something that's going to prevent or stop pain. And so examples of those, a big jump in spend. That's a great thing to alert on. Again, we can get in front of that when we have variable expense. So that's one thing we can get in front of. The second thing is renewal notice alerts, being able for us to tell you 90 days, 60 days, 30 days ahead of time that this renewal's coming up. So that's a really good way to get ahead of it. And the other one that we're really excited about is risk alerts. So we're seeing right now, again, with all these new AI tools, that companies are often changing their data privacy policies, and we can alert in real-time when those things are changing and really get in front of some pretty big issues if your data's being used in ways that you don't want it to be used. So I'd say recommend to the CFOs, they set up alerts that are, one, truly reducing pain, and then, two, just make sure they're accessible. So set them up where your employees are. Do they spend their day in email? Do they spend their day in Slack? Hiding them in a system really can't be the only place, and so just ensuring that you are where your team is and obviously putting them on repeat. Like I said, if you can alert 90 days, 60 days, 30 days, the more notice, the better.

Megan - 16:08: From your experience, what are the biggest challenges that CFOs face in vendor management, and how can technology and particularly AI help streamline the decision-making process and improve their outcomes?

Sara - 16:20: In terms of the first biggest challenge, we've heard from hundreds of CFOs that really accountability and transparency across their organizations is really one of the biggest challenges. They set budgets for their department leads at the start of the fiscal year, and then both parties are just really in the dark until someone runs a report, right? Or they spend hours putting it together, and they go around trying to hunt down the department leads to see how things are actually looking, and they spend days and days doing that. And, really, none of that needs to happen. So good software and AI can speed up decision-making with real-time data and visibility provided to all parties. Again, I think AI does a particularly good job here of taking data out of legacy systems—we'll say accounting systems in this case—and showcasing it in really useful ways. And that can be tracking budget utilization, alerting on overspend that we just talked about, or even, like what our customers love, is just holding department leaders accountable, but giving them the tools to actually do that and so that everybody's really on the same page. And when everyone's on the same page, obviously, time is saved, money is saved, and you just get a lot better decision-making.

Megan - 17:29: And when it comes to vendor relationships, how do you balance the need for cost-effective solutions with the need for quality, reliability, and innovation in contracts, and particularly when that comes to financial forecasting and vendor performance?

Sara - 17:45: I think that used to be an old concern, is how do you balance making sure you are a low-cost product with high quality, right? And

Megan - 17:55: Not always easy to do.

Sara - 17:57: Yes. Unfortunately, the good news, I'd say, is I think the cost of software and services will likely be coming down over the next decade as software becomes more affordable and efficient to build. I think we're in a pretty interesting time where companies can be both cost-efficient while using great solutions. With all that said, contract management, it must be reliable. It must be 100% accurate. The extractions we talked about, the data has to be secure. So if you're looking for vendors, you need to make sure the vendors invest here, and they stand by it publicly. So on the contract term extraction side, look for vendors that guarantee 100% accuracy in a short time frame. So I'll give an example. We saw a vendor recently that does contract extractions, but they do it manually. They do it over the course of weeks, and they do it with a team of individuals. That's not secure. It's not timely. You're basically just outsourcing. So look for vendors who can get 95% of the way there with AI and maybe just spot-check the rest. And then all vendors that have this level of access to your data, you know, they must be SOC 2 compliant or the equivalent. It's really easy nowadays to obviously see who has that level of security. You can see it in their FAQs or on their website. If not, you can just ask them. But it really is becoming a norm for vendors to have that level of compliance, and you should stand for nothing less. To answer your question, I think we are in an interesting time where maybe you can have both. You can be cost-effective and also compliant. And, obviously, StackPack certainly checks both those boxes.

Megan - 19:27: This is outside the list of questions, but I'm just wondering, do you have a customer story that you are particularly proud of? And if not, that's fine, but I'm putting you on the spot.

Sara - 19:37: Absolutely. We have a number of them, actually, but I'll just do a recent one here. We work with a Series B health tech company. They have, oh my gosh, so many vendors that they work with. And being in the health tech space, they have to be very, very thoughtful around compliance, around the use of data for PHI, for any type of health information. So they really care about vendor management. So they have implemented our tool. When they first started working with us, their biggest problem was that they were missing renewal dates left and right. So that has been our number one focus with them, and we received a note last week, actually, where they said, "Hey. You know, you guys just helped us a lot. You gave us 90 days' notice and then 60 days, and you gave us enough time to be thoughtful and to negotiate this partnership." And in the end, it got us to a much better place and actually, you know, saved us $7,000. Why we love that story is you look at the ROI, we're only $99 a month. So $1,200 a year. Think about that. That's just one vendor. We returned a seven-times ROI on that. So that's where we get really excited to be able to multiply across the entire vendor base.

Megan - 20:51: That's amazing. So is it $99 a month for everybody, anybody who wants to use it?

Sara - 20:56: It is. It's unlimited seats, unlimited vendors.

Megan - 20:59: Wow.

Sara - 20:59: So a really easy, cost-efficient, fast tool that you can onboard in just minutes.

Megan - 21:05: And with your background in tech-driven solutions, what advice would you give CFOs about integrating AI and automation into financial workflows? I mean, we can look specifically at vendor contracts and manual and minimizing manual intervention or more broadly.

Sara - 21:23: I think it would be really hard to do that on your own or using something like ChatGPT. Obviously, you can throw contracts in ChatGPT. I think you'll probably get a 10% accuracy rate. So I do think you need to find a partner, some type of AI-powered software that helps do this for them, something that's specialized. My recommendation would be to find a vendor that can do two things at once. This is what's a bit new. But ensure that they can do vendor and spend management, then you can marry that with contract management. For example, if you're just putting your contracts today in Google Drive or Dropbox, that's not particularly efficient. It gets a lot more powerful when you combine all those contracts with a vendor management system of record. Because you have those two data sets together, the whole system becomes smarter. So that if anything changes on the vendor level, it can then be changed on the contract and in the forecast level. So I'd just say look at partners that can help you leverage both of those data sets together, and I think you'll get a lot better decision-making. You'll feel a lot smarter. You'll have a lot better insights.

Megan - 22:30: And for CFOs out there listening today who may not realize that they need something like this, what are some typical pain points that they should be looking out for that might indicate that this is something necessary for them?

Sara - 22:46: I think what you can do is, and I'm sure, passively, CFOs are doing this now, but basically taking a pretty surgical approach to looking at your organization and the workflows, looking at and seeing where your colleagues are spending the most time on manual tasks. And I think the good news is there's a lot of AI solutions out there that have been created to help reduce that work and uplevel your teams. And so for vendor and contract management, we typically see, again, people just spending tons of time on getting approvals around the organization, chasing people down for contracts, missing renewal dates. These are really obvious places where you can throw some solution in, and it will do most of the work for you. So I think even just looking around and seeing where your team's pain points are and where they're complaining, honestly, is a great place to start.

Megan - 23:40: And looking ahead, what do you see as the future of AI and contract management, particularly for CFOs? Where is your own product heading?

Sara - 23:49: I think we're headed in a pretty far fun direction. Yeah. We see a world where discovering, evaluating, selecting, and managing vendors is just going to take a lot less time. So when you bring transparency into the system, which AI does, two parties, the company and the vendor, they have a lot less to dance around. So at the discovery and evaluation stage, AI can help a company quickly assess if the vendor's a fit or not. That's great for the company. It's great for the vendor. Everyone's going to spend a lot less time. At the negotiation stage, if a company's policies are set, the AI can just ensure that the contract aligns with those policies in minutes, not weeks and months like it normally takes to redline and negotiate a contract. Those cycles, they make it so quick that we actually find that most vendors are operating on the same template, like one simple contract. It makes no sense to have a different contract for every vendor. At the bottom of the life cycle with vendor and renewal management, AI can become eyes and ears to make sure that we're aligned with vendors strategically. So from a cost perspective, fit, compliance values, all of it aligns and ensures we're really working with the right partners. Every company has more vendors than employees, and we need to start treating those vendors more of an extension of companies, just like we do employees. And that begins by managing them.

Megan - 25:08: And, Sara, one last question for you. But as you look into the future, what is keeping you up at night?

Sara - 25:15: I'm sure people think that the obvious answer here is AI, but it truly doesn't keep me up at night. I think we found that AI, while incredible, it really isn't going to replace the job of a CFO, right, or an accounting manager. It's only going to really improve the way that we do work. It speeds up the tedious, and it leaves more time for strategic. So over time, it may reduce a headcount or two needed on the team. I do think a lot about how early career folks will come up through this system. Obviously, I think it will look a lot different than when you and I were coming up, which may be a good thing. Instead of doing tedious work for years, maybe we'll just start training out of the gate for strategic thinking and judgment. And, honestly, that's not such a bad thing. The thing that does keep me up at night is really just the pace of innovation. You know, we're making big strides on the AI front in really fast cycles where the technology, it feels materially more advanced every six months. And so how do CFOs keep up with that? How do their businesses keep up with that? We're expecting to see a big shift in software this decade. Businesses moving from legacy systems to new solutions. It's becoming easier to build AI solutions more than ever. Just think about how crowded the software landscape maps are and just imagine how much more noise there's going to be. So with all the new SaaS, new services to keep an eye on, I just think there's a growing need for guidance to help businesses find, assess, and use the best AI products for them. And how do we reduce the noise? How do we help companies use the right products at the right time on the right terms? That's what we're focusing the next decade on.

Megan - 26:52: Great answer. Thanks for that insight. Sara, thank you so much for being my guest today.

Sara - 26:57: Thank you. Thanks for having me.

Megan - 26:58: I really enjoyed speaking with you, and thanks for finding the time to be here with us today to share your experience and knowledge. I wish you and StackPack all the best, and congratulations. Sounds like you've created something truly powerful and wonderful.

Sara - 27:14: Yeah. We're excited to work with CFOs out there. Thank you.

Megan - 27:18: And to all of our listeners, please tune in next week. And until then, take care.