SaaStr AI Day Streaming Live: How AI Native Contract Agents Accelerate Revenue with Agree's CEO

Marty, the floor is yours.

All right, thank you. I'll go ahead and share my screen now. So yeah, quick introduction on my side. Marty Ringlein is, as I said, founder and CEO of GRE.com but just for some quick context and background setting for the rest of the talk. I've got about two decades in and I've seen it all from the AR AP automation side, operator side, been a founder a few times over and then had led AR AP Automations at Fortune 100 companies within the federal government and then even within early stage startups and then even within financial institutions. So I'll be talking about it from that context. I began earlier my big data career at the White House when I was working with the Office of Management and Budget, particularly in the procurement space around thinking about how we're using massive data sets and this is all about 15 years ago so you know, predates the current generation of gen AI and then more recently got acquired my last startup by Brex and then help build out their spend management system again all around ARAP Automations. And then I lead a venture capital fund which invests in early stage fintech and financial services companies and I teach a course at Northwestern University around new and emerging technologies, specifically generative AI and their impacts on the P and l. And then agree.com is tooling all around native contracts and then agentic revenue, which is the topic we're going to jump into. So one of the things I know that we're all, you know, AI has taken the sort of breath out of the room, it's all the rage. It's everything we talk about. We have entire conferences like Saster AI Day dedicated to it. So we're all pretty well caught up on AI. I think we all at this point are pretty familiar with AI agents and now we're entering into this agentic era and I think some of us, a few of us know what that means, have dabbled into it, but this is moving quick and it's moving fast and it's really a about workflows. And so when we say agentic I think we're really talking about where does a pre existing workflow exist within an organization and then what are the opportunities for that workflow to be more autonomous and sort of be more operationalized in using AI to do it. And so one of the places that hits home closest to all of us is on the revenue side and as we think about our revenue stacks and so this is already happening in real time, we're already seeing companies deployed these measures where AI agents are already, they're creating invoices, they're processing payments, they're collecting lost revenue. And the impacts are pretty staggering. And you know, I think that's, that's why we're here. And we want to understand how those impacts can, can, can come and directly impact our own business. But just some of the quick stats that are up on the screen here, 32% on industry average increase in same day payments or we're even going to talk about the collection of payment data which is almost as valuable as Getting the payment 4.3x on overdue invoices recovered. So on the, on the revenue recovery side, and then one of the biggest ones that we don't talk about enough is just the amount of reduction on the human in the loop side with human errors, manual entry and just the efficiency and savings that happen there. And I think the biggest thing here is that this, this era, it's already here and it might not happen within your organization. You yourself or your team might not be ready for it. But one of the things we talk about, especially at Northwestern, at Kellogg, is it's not really a question of if you're ready for it or if your team's ready for it. The bigger question here is what if your competitor is ready for it? What are they doing with these tools? How are they moving quicker, moving faster? How are they recognizing revenue in a more efficient way? And then are we losing by hesitating here and not moving or delving or learning or exploring quick enough? And that's where sort of the opportunity exists. So agentic revenue, I'll do the thing you're never supposed to do and just kind of read from the slide. Just because I think on this particular slide it's important. We're talking about a new construct, something that, a term that we may have heard, definitely agentic commerce hit the scene pretty hard and heavy just about five days ago. So this is pretty timely where OpenAI comes out with stripe and reduces this new protocol all around the future of agents doing more commerce on the web for us and I think the predictions are over the next five to 10 years we're going to see more agent activity on the Internet than we do human activity. And that'll be a fundamental shift in how we all do business. But basically we're talking about revenue that flows automatically from contract to cash and then all of that being powered by intelligent agents. And we'll talk a little bit more about that, what it means, but this is that AI agents managing that life cycle of a contract rather than relying on sort of the human in the loop process that most of us, us have today. And this, this is a transformative moment. This is what I'd call a shift, I think, you know, a tectonic shift at that. Especially within our day to day where the concept of lifestyle management I think is changing. So I think we've always thought about a revenue in terms of a revenue life cycle. I think we've always thought about contracts in terms of contract life cycle management. We even call them CLMs. But now this is more about workflow automation. So think about your revenue less so as a life cycle and more so as a workflow. How does it go from point A to point B? What are all the steps along the way, all the humans and all the steps along the way and then what are the opportunities for those to become, to have more automations and for those to become more autonomous where they're happening sort of on their own and when that happens more quickly. So I think we want to think through in terms of all of our revenue, where are the workflows, all of our workflows, what are the opportunities for them to become agentic workflows? And that's really what we're talking about when we say agentic revenue and entering this kind of new era. So here's the problem. 70% of B2B revenue flows through a sales contract. That's why we're here, that's what we're talking about, that's why you're interested. But the contract to invoice, to payment lifecycle that most of us use today, it's slow, it's manual, it's error prone and worst off. And this is the one that we definitely don't talk about. It's siloed. The current process today for most of us is, is somebody traditionally on the sales side will go out, they will find a customer, they'll do something like inbound or outbound sales efforts to do that. They'll do a demo, they'll have a conversation, they may send a quote or proposal, but ultimately they're going to get a contract in front of them. That contract's going to get signed. That signed, executed contract goes from that sales rep from the sales side. It now moves over to a different silo on the finance side. And that's usually a manual process where that gets handed over to somebody by a email. Slack, it might even be the swivel of a chair and saying hey, close the deal in the ringing of a gong. And then somebody from finance Goes and typically grabs all the information from the contract manually and puts it into an invoice system like a Bill.com or QuickBooks and creates an invoice, schedules that invoice, sends that invoice and then we wait the net 45, net 60 for the customer to pay. And then when they don't, somebody from finance goes to somebody in sales and says, hey, go track down this invoice. And now it's sales has to play the role of collections, which is a tedious job. But more importantly, every second that somebody on sales isn't closing a sale is a missed opportunity, not only on the commission side, but on the revenue generation side. And so these are just huge opportunities for us to close these gaps. And so when we think about this new era of AI, especially in the context of our own revenue and this, this, this agentic revenue, it's how do we close those gaps? Because this is a con emergence moment. It's, you know, I think the most dangerous thing that you can do in business. When you ask, you know, why does it work this way? Why does it work the way that Marty just walked through? Why, why do we have these silos? Why is there this separation between legal sales and finance? The most dangerous thing that we can say is, well, it works that way because that's the way it's always been done. And I think what we're seeing now in this new era of AI is an ability to question that status quo and to rethink some of those processes because those were life's cycle processes and now we workflow automations that we can put into place. And that's, that's going to be a convergence, not only convergence of technology and our business operations, but a convergence of our teams where we'll see just holistically on the rest of the organization. The engineers will do a little bit more design, the designers will do a little bit more engineering. Well, the sales team will be equipped to do a little bit more on the legal side and the finance team will be equipped to do a little bit more on the sales side. And you'll have these worlds coming together which creates massive opportunities for all of us. So I wanted to show this slide just really quickly because we're, you know, this is Saster AI day. For everyone that doesn't know Jason Lemkin, started EchoSign, sold it to Adobe 14 years ago and you know, agree is a modern day version of Echosign and it's a modern day version built in sort of the AI era. The agentic era. But what I thought was funny when I first saw this, because this speaks to not only the crux of why this talk even exists, but you know, why, why we even started agree because the sign at the time infuriated me so much. It's like, oh, I have to nothing I can do as a founder but go and build the solution. Because when I saw this, it said send sign done. And it's a really great marketing slogan because it speaks to it being simple, easy to understand, easy to use. So you know, it's a really great value prop. But from a sales perspective, what really bothered me is we know that it's not done when it's signed because we all know the deal's not done until the wire hits. So this becomes an important moment where we have to ask ourselves why don't our tools work that way? Why are they not done right at signing? Why don't our teams work that way? Why doesn't our data work that way? So we know it's not send sign done. We know that there's still many processes within that workflow, within that life cycle. But we have these tools and we think about our data and we think about our workflows or automations in a very linear sort of handoff perspective. And that's not really true anymore. And so I want us to really think through the fact that it isn't send sign done. There's really a lot more to it because we all want to grab the revenue, whether it's to get the commission or just to move the business forward. And how do we do that more effectively? So here's the secret sauce, pretty obvious I think at this point where we start to converge these different aspects within the workflow, how do we bring our sales contract, our invoicing, our payment processing and even our revenue recovery together and then when we do that, how do we start to see the net value of that? In this case, you know, what we're seeing on industry whole is speeding up revenue by over 200% and a majority of that is bringing these traditionally siloed teams a little closer together through some of these workflow automations. And so to do this, what we're really talking about is that we have to fundamentally believe that the contract is the source of truth to revenue for the business. Now again, I'm talking primarily to sales led organizations where a majority of their revenue is flowing through a contract and that's a sales contract. And we have to remember that at the end of every signature somebody has to pay somebody Money. But the process to do so has traditionally been pretty disjointed and broken, where we have somebody doing the signature, somebody doing the invoicing, somebody doing the revenue recovery. Now we're going to start to bring those together. And if we do it all around the contract being that source of truth, of revenue, it changes things, especially from an AI perspective. And so I'm going to kind of keep coming back to that. But we like to always say that the signature, again, the deal's not done when the contract signed, the signature is really just an invitation to be invoiced. And I want to think about it that way. So remember, everything starts with a contract. Of course, there's a sales pitch that happens. There's an inbound or outbound effort that occurs, there's a demo, there's a qualifying function, there might be a quote or a proposal, but it's not real. We don't really kick off the life cycle or the workflow until that contract gets signed. And so once that happens, something magical happens. Because I like to remind us that AI is amazing as it is, and as many opportunities as it affords us now and all the potentials in the future, it's a little limiting in that it's a lot like the telephone. When the telephone's first invented. One telephone is the most useless piece of technology ever invented. The telephone itself has no value until there's a second telephone. And so AI is a lot like this. Without data, AI is absolutely incredibly useless as a technology. But with data, and with massive amounts of data, it becomes this, this game changer that we're all starting to recognize it as now. And we take the position if the contract truly is the source of truth to revenue for the entire organization, then AI can be fueled by all of the data in that contract. Because what makes a contract unique, specifically a sales contract, is it's full of invaluable payment data. We're talking things like how much somebody has to pay the payment amount, who has to pay who, when they have to pay the payment terms, when it's due, if there's a discount rate, what the scope of work might be, all of these attributes. Even if there's an escrow, what holds the escrow, what releases the escrow, if there's a deposit, the terms around the deposit. All of this data lives within that contract. But that data, those are just variables. And it turns out that they're the exact same variables you need to generate an invoice or a purchase order. So if you had the Contract. And you thought about the contract as this fuel to AI, and you thought about it as a contract filled with payment data. Now you can take all those variables and instantly, dynamically, generatively create an invoice. We don't need that human in the loop. We don't need that manual entry, and we don't need to be exposed to that human error that often occurs with manual creation of invoices. We can create those instantaneously because that data pre exists somewhere. We've just never closed the gap between the tool we use to execute the contract and the tool that we use to create the invoice. And now that we can start to do that and use technology to do it, we can move quicker, faster. Because the one thing that I love to sort of talk to really quickly is that we love E signature because we could still print a contract, sign it, send it an envelope, and use a pen and paper. But the problem with that is it takes too long. So if I send something for E Signature, it can be sent quickly. Why don't we have that same sort of mental model around invoicing? Why does somebody from sales send the contract fully executed, to somebody in finance who manually creates the invoice and manually sends it manually follows up? That process also takes one to three days. And we're okay with that, but we're not okay with it on the E sign side. So we like to say, what if the invoice could move as fast as the digital signature did? That would be an absolutely incredible sort of step forward in terms of efficiency. And the last piece in all of this is if you have all the contract data, all that payment data, and it is used to facilitate the creation of the invoice. When you take contract data and payment data and you put it together, then all of a sudden, at the end of execution of a contract, you don't need the little interface with a little checkbox that says you signed, which is completely useless utility for us. Why wouldn't I take you directly to a payment page where you could do a number of things? One, it'd be great if you paid, because for a number of our clients, they're incentivized to pay quickly because they want what's on the other side of that contract. They want to lock in the vendor, close the schedule, take the property off of the market, you know, any number. They want the SOC2 compliance right now, so they want to pay for those that don't. And they do want to wait the net 45 we can schedule a payment and so because now we're in what we call buyer intent, right? The customer, the buyer has gone through the contract, they reviewed it, make sure it lines up with everything we talked about, and then they've connected the dots to say, hey, there's an opportunity here where I'm already in the mindset, I'm already in the mode, why don't I put that data in and facilitate the scheduling of this payment and that moves things quicker for us. And then the recovery side, this one's kind of obvious where now that we have all of the terms of the invoice, we know when the Invoice is due, net 45. So we can remind you a week before it's due, we can remind you when it's due, we can remind you when it's a week past due, but we can also reference the legal terms and legal obligations of the executed agreement. So it's more than just being overdue now. It's about being sort of in alignment or not being in breach of the legal document that you sign. And that's a much more important, much more impactful document when it comes to making sure that we've facilitated the payment. Now again, I already kind of referenced these stats, but just to hit it back home. 32% same day payments, 4.3 times overdue, invoices recovered, reduction in manual entry up to 70%. Like these are, are big and these are things that you can't ignore. There's a quick model for this and this is something that we teach the at Northwestern. But essentially all you really want to do is look at your current workflow, find the gaps that are in the workflow and really look for the data gaps. I really want you to think about this less as humans moving between humans because that's how we're used to the life cycle. Now think about it as a workflow and an agentic workflow where it's data moving throughout the workflow and then where are the gaps in that and then how can we automate that, how can we make that more autonomous and then do things like measure it to make sure that everything that I'm saying here actually comes to fruition and that you're seeing those net positive results. I'm just going to end on this quote because I think it's perfect. It's from Jason. So I started with Jason, I'm going to end with Jason. But he said this recently. By the end of Q3, which is now, because we're in Q4, we'll have 10 distinct AI agents running in production at Saster, not as a tech experiment or marketing stunt, but as core members of our revenue and operations team. Now, I think that's, that's already true at saster and that's 100% going to be true at all of your businesses if it's not already. It's going to be, you know, it might be within the next two weeks, the next two months, the next two years, but this is one where it's an inevitability and then that's all I have for you. And I just want to remind us that when we talk about the sales and finance alignment, the bridging of these two teams, which ultimately that's what this is and that's how we speed things up. Finance wants to recognize revenue faster. Sales generally wants to see their commission quicker. Those are the exact same things. It's just a vernacular change, it's just a vocabulary shift. But we're all after the same end game. All right, so now we're moving on to Q A.

Awesome. Thanks, Marty. We do have a question in chat. How does the model you describe work in professional services where you have a variable scope of work?

Yeah. So, you know, there's a number of ways to look at this and, you know, one I might use, you know, a common term now, usage based billing. But, you know, whether it's on a schedule or it's recurring, I think the most important thing is how do we kick things off. I think most of us know from the sales side, getting to the signature or getting to yes is the biggest thing. Right. Then sending over the contract, getting them to signs the next sort of hurdle and then getting that first payment down is sort of the next big one. And that's why again, I want to bring this back to a data conversation. Think about it less in terms of, hey, how do we get that revenue? Because I know that's how we're all kind of our mental models are trained. I want to think about it more as how do I get their payment data stored into our system to do any of number of things in the future? And I want to get there as fast as possible. So yes, let's still get to yes, as fast as possible. Let's still get signature, velocity and signature down on contract as fast as possible. Now let's get payment data, whether that's they're connecting their bank, their ach, their wire information, talking about their credit card information. Let's get that into the system. Whether it's to kick off the deposit, whether it's to kick off the first invoice and then let's make it it easier for all of us to work together going forward. So if it is a continued relationship or it is a usage based billing relationship, we've already got their details. Now it's just automating that workflow of how do we know what the client owes? Because every recurring invoice or every invoice going forward will be a different price. But the most important thing is making sure that we can build them quickly in automatically if they, if they so choose or get the invoice to them if they want to do it themselves manually.

Great. We do have another question. Here it is. Where does the CPQ process fit? Where CPQ equals configure price quote.

Yeah, so you know, I talked about that a little bit. Where we have there are moments in the flow that predate the execution of the contract and that would be something like a quote, a proposal, you know, it might even be an old handshake agreement on a golf course. But we want to capture that data and then to the best of our ability. But that data constantly is in flux and in changes. So still capture everything in your workflow. Right. And so in your case the CPQ is a very important part of the workflow because you're sounds like you're always sending a quote prior to the contract. Sometimes the quote might be the contract. You know, different use cases, treat it differently. But, but again think about it from a data perspective. Figure out where you're using. So if we've got an all in one solution, I'm not here to plug Agree, but like Agree where it's contract invoicing and payments all in one, that makes it a little simpler. But if you've got a disparate system where you've got one tool for invoicing, one tool for contract execution and one tool for quoting, just make sure that you're tying those together. So back in the model that I shared in the. Down that process because for me the quote is less the important thing because that's still getting to yes. So when I say getting to yes, it's still the hey, we agree on the quote, it's that signature down. And so again they might sign the quote and you might treat that as your contract. But signature velocity, I want signature next in the flow. So get to yes quickly, get to signature quickly, then get to payment data quickly. Whether that's the first payment or just storing the payment for the first payment later.

Love that. Our next question is does Agree.com work with finance companies to get invoices paid faster. For example, supply chain or factoring companies.

Yeah, and so this is one of the big aha moments where you know, at the end of the day a group. Does the invoicing. So it's kind of a rip and replace of a Bill.com or QuickBooks invoicing. And then it also does the payment and then it does the collections. And so, you know, we'll, we'll call it intelligent dunning, but for us it's how are we using the data to go collect more of our revenue faster? And so one of the biggest things which seems so obvious is let's just right after execution, when they're in that buyer intent, when they're in that moment, let's get them to the payment page as quickly as possible. Even if they're not ready to pay, let's get them to schedule the payment the same way they would use in their own sort of ARAP system. So let's get them there and then let's go one step further where we can use the, both the contract data and the invoicing data to go and recover those funds more quickly, more efficiently. And that's a number of ways because right now your invoicing tool like, like a QuickBooks, a bill, even, even a stripe invoicing, they don't know what happens when the contract's late. More importantly, they don't even know if you don't have terms in your contract if it's late. Right. Which you should all have. And most of you on the mid market will have this. Most of you on the SMB side won't have, have this. I think it's 70% of SMBs do not have any kind of terms within their sales contract for what happens if a payment is late or missed. But when you have those, you can reference those and that just becomes a stronger hey, your payment's delayed. Hey, your payment's late. Those are easy emails to ignore. But the eventual hey, you are now in breach of, you know, the, the, the, the legal document we signed, the terms of those are now in breach and that's going to be void. If we don't rectify that, that lands a little bit stronger, but it does. Another thing, the person that signed is often different than the person that paid. When we can remind the person that signed that because the team at their company who pays has not paid on time that there is a breach in the thing we signed together, you, you end up evangelizing somebody on the inside to walk over to the finance team and say, hey, what's going on here? Let's speed this up. And you've got an inside person that helps you, and that's probably the most valuable thing. And then one that's just anecdotal that I think, I think, silly, that most invoicing companies don't do this today. And again, most of these invoicing tools are 20, 22 years old. So we're just in a new era. But they don't simply tell you like in the audit trail for e signature, when the contract, when the invoice has been opened. Because most of our clients, the reason they didn't pay yet is they tell you, well, you didn't invoice me. And you're like, well, I absolutely invoiced you 45 days ago. But it's set in spam. And we just never looked in the logs to say, hey, that email was never even opened. And so that's the first reminder is I don't want a payment reminder to my customer that they're late. I want a payment reminder to my team that, hey, it was never open. We should do something about that. Let's make sure it didn't get stuck in the promotions folder, the spams folder. And the same thing is true on the sales side with knowing when the client paid, when you combine all these things together and you have payment data. Most sales teams don't have access to the books. They can't see the bank account or the ledger of the company. So they don't know when, when the wire did hit. Right? The deal's not done till the wire hits, till the cash, til the ch clears. But most sales teams don't know that. They have to wait for someone to tell them. Imagine that data being pulled back into an audit trail where I can see when my client signed, when my client opened that invoice, when the, when the ACH was started, when it cleared, when it processed, and when it hit the bank. And then we can get started on our project together. It makes the client relationship stronger.

Amazing. And we have one more question, but we're right at the top of the hour, which is where we're technically supposed to stop. I'm going to read the question, Marty. Think you can answer it in maybe one or two minutes or less? Then let's go for it. If not, then maybe we'll just follow up afterwards. There's been a consistent message today about the importance of training AI agents effectively in the context of agentic revenue. What kind of training Data or human feedback loop matters most to make these contract agents accurate and trustworthy.

Yeah, and this is one where, you know, if we're talking about the AI hype cycle, you know, let's, let's go into it with a bit of caution. I think we'd be very optimistic about where this technology is sending us, but let's be, let's, let's go in with some caution. You know, I'm not suggesting that we're going to go full autonomous here. For example, I bet most of your finance teams, they're not comfortable with AI sending the invoice, they're very comfortable with AI generating the invoice. But I bet they're going to want to get it a one, a once look over. And so that's still what's happening today. And I think that's going to persist over the Next, call it 24 months. And again, the data gets better, the models get better, the accuracy gets better, but it's still pretty performing at better accuracy and performance than when the human was in the loop. The human error is still the highest amount of error, even with there not being 100% accuracy on the LLM side. So. Yes. Are there hallucinations? Yes. Are there sometimes where something gets missed? Absolutely. Is it happening at a magnitude less than when a human does it? That's the important part. And the answer is like undoubtedly across the board, yes. So look at it from that perspective. But use your historicals. You've been in business for 10 years. You've got 10 years of contract data, you got 10 years of invoicing, you've got 10 years of customer data. If you can pull those historicals from your current vendors and use them to make sure that you're optimizing any kind of models or partners you're working with in the future, you'll be able to do kind of what we'll call like a digital twin analysis where you can look at what had happened and what we predicted on past behavior. And if there's a level of accuracy there that you feel comfortable with, then you know that going forward you're going to be a pretty good spot. But you know, again, being, being a technology optimist, I'm still, you know, give it a once look over before you send the invoice, give it a once look over before you do anything, just to, to spot check it. But you'll still move quicker, faster than when it was just all human led all the time in silos.

Amazing. Thank you, Marty. Technology optimist. I love that that's my new thing. Everyone, that's going to be a wrap on this session. Marty, again, thank you on behalf of everybody at Saster for doing the session with us today. Our next session is starting now. Flat Files VP of AI is starting their session. The Zoom link is on the website. And as I mentioned before, this recording will be posted on the Saster YouTube channel as soon as we're done with the live stream. Thanks again everybody for joining. Marty, great to see you. And we're.