AI is rewriting revenue for finance teams

Hi, everybody. I am Amy Connery, and I am joined here by Michael Mansard and Doug Patterson, and we’re really excited to be talking with you today about all things AI and finance, how AI is rewriting revenue, creating all kinds of challenges and opportunities, and we’ll provide some frameworks, some guidance, and some actionable leave-behinds that you can take back to your organization. So we’ve got a lot of great content to cover, and I’m going to lay out really quickly what we’re gonna talk about. First off, we’re gonna do some stage setting, the squeeze, costs are rising. We’ll talk about, like, what is it that you are facing within your organization? Just a little bit of stage setting to get us all in the

same page. I’m sure we’re all contact switching from meeting to meeting here. Uh, then we’re gonna talk about solving monetization. Michael’s gonna lay it all out. He’s brought some great frameworks and models that are gonna help you demystify AI monetization. And then we’re gonna talk about agentic finance, why you need it. Hint, hint. You need it because of monetization, and we’ll talk a little bit about that. And then finally, we’re gonna talk about how to enable the flywheel and what that is. So without further ado, let’s do some stage setting today. So I’ve been around the SaaS space for a long time, and one of the things that we’re seeing happen right now that is absolutely unprecedented is that AI is breaking SaaS

economics as we know it. Uh, it’s a bit of death by a thousand cuts. It’s been happening slowly, and all of a sudden, all at once. For a very long time, SaaS economics meant that your heavier user customers were more profitable, and we’re seeing AI invert that equation for a few reasons. Uh, first of all, obviously, you have costs that you a- incur associated with heavier use. Uh, we’re also seeing some of the metrics break down, and gross margins are really shifting with AI, uh, somewhere around the lines of, uh, Michael’s research found about a 20 point shift in gross margins from pre-AI to, to today what we’re seeing. Um, and so we’re absolutely seeing even McKinsey

is, is saying that AI-driven margin compression is one of the force h- four horsemen driving a lot of what we’re seeing, uh, as being dubbed the SaaSpocalypse, right? The valuation decreases. We’re in a very uncertain time, and, uh, and, and we believe and will talk through how companies are managing their way through this, but it’s absolutely something that we’re all feeling today. And so you and your finance organizations probably have a lot of tools and techniques that you’ve used in the past to manage margins. So if you think of the, the historic price waterfall, whether or not you own pricing within your finance function, and many of them do, uh, you have this price waterfall,

and in the past, the way that you would really take control of margins is by looking at negotiations, so contract terms in B2B SaaS, volume discounts, payment terms. These were ways of holding line on margin. And sometimes when a sales rep might wanna do a deeper discount, it would have to get an exception level, and that’s how you managed it. Now, uh, if you are including AI within your solution, within your product that you’re bringing to market, you’re now thinking about token bloat, uh, inefficient prompts. You know, how might your customers be using your tool that is causing you to incur costs? They’re experimenting. It’s early day- early days, but every single one of those wasted tokens

or those early experimental tokens that your customers are, are using as they’re experimenting with your features are creating real cost for you. And sometimes those costs might be a surprise, right? It’s hard to know, uh, what the different costs of the LLMs might be day to day. These are always shifting. There’s hypotheses that these costs are gonna go down over time, but no one really knows. And so in addition to this lower margin, there’s also this element of unpredictability, as well as these engineered ways of protecting margin that are new for finance organizations. And finally, the third thing that is happening in the squeeze is this value equation is really shifting. So SaaS economics being broken by the differences in margin

as well as in the fundamental atomic unit of how you measure the value of a SaaS application in the past. In the past, applications typically equated with users, which typically equated with seats. However, we know in the age of AI, the work being done by applications isn’t always tied to a user anymore, and it isn’t just because maybe your customers have less users than they used to. It’s that more and more they have agents, uh, and you have agents that will be doing the work that a user might have done in the past. So whether it’s resolving tickets, whether it’s processing data, all of these things are being done increasingly by, by algorithms, by agents, and not as much by users.

So, you know, if you’re paying attention at all to pricing in the SaaS market, you know that this is an issue, and you also know that the market is moving, but you may not realize how quickly it is moving. So a couple years ago today, the primary pricing metric was, in fact, seats, and about 18% of companies, and this is Michael’s research that he’ll talk about, were doing some sort of hybrid, some seats and some usage or consumption. Today, it’s almost split a third, a third, a third, so seats, hybrid, and usage, with usage obviously a growing proportion, but the prediction is that hybrid is actually gonna be the category that’s going to grow more and more. It’s what we see, uh, as best practice when we do the analysis

within the institute in terms of how companies are growing, and Michael is going to get into more detail about that very shortly. But the bottom line is that, uh, when it comes to solving for margins, when it comes to solving for the value equation, companies are at different stagesAnd I think the last thing I wanna make sure I, I put out as far as the squeeze, you’re likely also facing a situation where you’re putting AI functionality out into the market, getting it into your customers hands. You want them to experiment and to adopt, but your opportunity to monetize that will likely lag that adoption. It’s just like with any new tool, any new feature, and, and we’re all experiencing that in our lives with LLMs, right? You’ve gotta seed

that demand before you can capture and monetize that. So, that’s also part of the moment that we’re in, and all of these different monetization mom- uh, methods that Michael is going to be talking about shortly, there are different roles that they play depending on the maturity of your customer with your tool, depending on what your tool does for your customer, and it’s really, really important to keep that in mind. As you are working through a maturity curve, your customers are working through a maturity curve as well, and meeting that moment requires levels of agility and flexibility, uh, beyond which we have not seen before. So, to talk about how to solve for all of these forces that I’ve just laid out, I’m really excited to bring in my colleague,

Michael Mansard, who has studied AI monetization, has done a lot of research, talked to many, many companies. And, look, the squeeze is real, but there are some companies that have gone forward into a moneti- AI monetization. We’ve got some lessons learned. We’ve got a trail of what works, and, um, Michael, I’d love for you to talk through what companies are facing and how they’re solving for this. Thanks, Amy. So, you know, the honest picture, and I think you painted it, right? Is that until very recently most companies were, you know, still pricing, uh, AI like it’s SaaS, right? So, on seats. Uh, they’ve incorporated, you know, AI on the product suites, they’re building agents, customers adopting feature, uh, but the pricing hasn’t really caught up. Uh, the good news

is that it changed very recently. So, of course there’s always been some companies ahead, let’s say of the, of, of, of the, of the pack, but right now everyone’s moving. Uh, what we see right now is that most companies are layering users with AI-specific metric or AI credits when not charging AI directly. So, uh, let me start just very briefly with a classic example that I would reuse. Uh, agentic AI in the customer service space. You know, you used to buy software for your human agents so they would be more efficient, but now, with agentic AI, you know, the AI does the work. So, most companies in that space, uh, you know, think of Zendesk for example, they’re pricing their AI per resolution, per solved ticket. Uh, Zendesk charges $2 per resolution I believe, or something like that. So, you know, while it seems very

attractive, uh, that’s not what we see in the markets. By the way, we’ll see it, that’s 10% of the markets. So, you know, why? Because there was preexisting budgets. You know, there was business process outsourcing, so people buy this way, the buyer were educated, there was budget you could tap into, and you could compare to something. And that, you know, is only a tiny, teeny ex- you know, number of companies that can really do that with AI. So that idea was the, the starting point of, uh, the research I started three years ago that you mentioned, Amy, which is the compass framework. Um, so we’ll be sharing most likely the articles associated with it. But what does it mean? It means that AI has created four fund- four fundamental ways to actually charge or to, to decide what the metric that you charge by looks like if it’s not a user. And each of them have different risk profiles and

different operational, uh, demands if you will. So, think of it as a spectrum, right? It’s not, it’s a bit blurry, but it’s a spectrum, um, e- which aligns towards value creation. Uh, so, you know, on the left side, you actually have per agent, almost charging AI like it’s an employee salary. It’s easy to understand, fixed fee for AI worker. It’s predictable. It’s, you know, again, simple, but it’s disconnected from value. And obviously, it bears the same risk that user-based pricing that you mentioned, Amy, depending on how the AI’s, uh, you know, the AI agent consumes, um, you know, the underlying AI infrastructure. It’s 16% of offers today. Then you can move to per activity. Think of it like, uh, an hourly consultant, so you can pay per compute unit, compute hours, token.

So, it scales with usage, meaning it covers your cost to some extent, uh, but time and effort are not always, uh, or often correlated with value. So, that’s one of the biggest bucket right now. It’s about thir- 37% of offers. And then, actually activities can generate outputs, which is a tangible delivery. Think of it like a BPO, business process outsourcing contract. You pay per action, per, uh, deliverable, per document, per simulation. Um, so it’s closer to value. It’s harder to define uniformly, but that’s also 37%. So, the big num- the big chunk of companies are really per activity, per output. And the last one, as I said, per outcome is only for a very few, despite the headlines it made, I would say, last year or two years ago. Excellent. And, and, uh, I, I think that’s a really,

really important to look at this distribution for people who are on this call. I think you’ve done a great job of helping sort out maybe some of the market hype from the reality itself. So, let’s talk through, how does a company decide where they, where they end up? And is it a set it and forget it exercise? Or is this something that is gonna happen continuously now? Well, first off, I wish I had the magic wand and the solution to actually solve for that problem, but I did not find it. Uh, what I found however was the fact that there’s this sequence, and going back to your question, Amy, that sequence is iterative and it continue. So, you probably mentioned it, Amy, as well, but I’ve seen, at least on my personal end, more changes in three years than in the past 20 years. So, you know, it’s iterative. It’s not gonna end and, you know, every single technological

breakthrough and customer understanding and adoption is gonna change that. That being said, um, you know, um, for the, the sake of this, of this webinar that we’re gonna try to, to keep as short as possible, uh, we could boil it down into four sequences. Um-The first one, you know, is how do you actually find, across that spectrum, the right metric for you at a given stage? Right? We said we were gonna iterate. Second, how do you zero in on the right pricing metric using, uh, the metric evaluator? Third, uh, do you, do you know, by the way, and should credit be a good model for you? There’s a lot of hype, as you know, around credits. They’re making a big comeback, but, you know, we need to, again, slow the hype from reality. So there’s a tool for that. And last, if credits are a fit for you, then how should you architecture them? And we’re gonna be using what’s called the 12 Attribute Blueprints,

which I’ll think of it as the Lego bricks of credits. Amazing. So let’s get started then, in that sequence, and I’d love for you to talk us through the, the matrix. Okay, so step one, that’s, that’s three by three matrix. So essentially, to keep it super simple, ’cause we could spend, uh, a half an hour on it, think of it as a triage, right, to decide what kind of metric is for you. So there’s two dimension. Uh, one of the dimension is the scope of the agent’s work, right? Is your agent automating tasks? Is it overseeing processes, or actually is it almost like managing entire goals, uh, and achieving the goals, uh, autonomously? So it’s almost a level of autonomy, in a sense. The second dimension is the level of attribution. You know, how, how much, uh, how clear is it for you and your client

that the value you claim you’re creating, um, you know, is actually accepted by both parties? You know, so is it, uh, indirect contribution? Is it a direct, so you have partial credit for the value you create? Or is there very clear and high correlation and causation, uh, between the value you create and the customer? And you’re supposed to be looking at a given time, right? It can be for a segment, by the way, uh, of, of your offer, where you fit, and that should inform you about which, uh, metric typology from the one I just described before you should be preferring. And you’ve got an example of Zendesk here, and what’s interesting is they have two different types of metrics. So talk about, uh, you know, any given company might find themselves in different places on this, this matrix- Abso-… depending on the products. Yeah. Absolutely. And Amy, to… So you… The Zendesk example is, is clear.

There’s plenty of other examples, by the way, on the market right now, and what you’re describing, Amy, can, can be true for segments. So you could have two different metrics to address the different segments, or it could just be that you have different types of actions or different type of agents, and you could actually use two different metrics to actually price two different results, uh, if you will. Got it. All right. So talk a little bit about, you know, given those metrics, you just… Zendesk was in that ru- l- like, maybe give me- Yeah…. maybe give me an example of how other companies are doing that. Uh, absolutely. So, s- s- so again, you know, uh, I, I’m sure, uh, most of the people connecting, you know, uh, are, as well, paying probably, right? A lot of finance folks in the room, so, so (laughs) you know, signing those contracts. But you’ve probably seen Box. You know, some of their AI capabilities are actually on the pay-as-you-go or on a credit-based model,

so you pay AI per operation, which is layered on top of your existing plans. Uh, PagerDuty, I think they have, uh, ramping commitments, contracted AI adoption, right, curves, uh, and you have, uh, ramping events, uh, across multiple years, so kind of like, um, uh, activity/output based model. Uh, I think Invoca actually charges per qualified lead conversation, uh, because, you know, it’s kind of like sales-orientated, so, you know, clearer to demonstrate, you’re closer to the end result and the customer. So there’s many, many examples. Um, we, we could spend, uh, a lot of time, and you can see, like, dozens of example in the Compass framework, and I’m sure you do have these examples in real life. Great. So you’ve picked your metric. Yup. How do you actually evaluate and, and zero in on that? What are the questions you need to ask yourself? To your point, Amy, think of Compass, right? Compass gives

you the, uh, kind of like the big direction, but then you need to kind of like refine on how you navigate. So once you know the model, your preferred model or your preferred mode- m- metric typology, then you need to stress test against that. So that’s where you have what I call the, you know, the seven dimensions, and each dimension has two questions. So we’re now gonna go through 14 questions, as you can imagine now. There’s a selection of questions on the right side of the screen, but essentially, you know, these questions, uh, make sure that the metric is fair for you and is fair and logical for your customer. So you’re supposed to score it against, uh, thes- these two points of views. Let’s take Zendesk as an example to go through some questions. You know, um, let’s imagine they could have charged per token, right, because the agent consumes token, but they decided to charge per ticket resolved for all the reasons we said.

Look, per resolution, actually score high on value alignment and acceptability, especially because the buyer were educated and they understand. Uh, token, actually score probably higher when it comes to cost pass through, because if you have, uh, you know, customers who has many, many, many questions and the resolution is long, then the risk, uh, is actually on, on Zendesk. But when you look at the 14 questions, it’s very clear, uh, that actually, um, you know, uh, the per resolution is a good model. Great. And one thing I wanna call on, if you’re listening to this and you just had a loop in your mind, like, “Wait a minute, where, what is this evaluator?” These are… You will be getting these after this call. So these are real frameworks and models that you can access today. Um, it is, it’s simply a set of questions to ask yourself, to, to, to give to your team

who’s making these decisions to help you understand that. So it’s, it’s very, very simple to go through. Uh, absolutely. Uh, Amy, if I try to summarize it very quickly, why does it exist? Because the first tool gave you a, a rough direction. Then you need to find a metric. I don’t have a tool to tell you what’s the right metric, but you can use that to actually benchmark them. And it’s supposed to help you avoid any blind spots, ’cause there’s so many risk when you pick a metric. So th- think of it as de-risking and making sure that it’s customer-centric and it also cover your own risks, own risks. So Michael, let’s talk credits. And before you get into the decision process of whether or not you even need credits, can you help me understand, what is the difference between a credit and a token? Because I hear about both. So it’s, so it’s interesting. So the first thing I wanna say, which is

important, by the way, is that credits are making a big comeback. Adobe, Salesforce, ServiceNow, HubSpot, OpenAI, Microsoft, you know, everybody seems to be launching them. Some they did before, right? Adobe was a very early adopter. And it’s roughly 30% of n- AI offers today which are monetized with credits in some shape or form. So going back to your question, the difference between credit and tokens, and I’m sure you have your own spin, Amy, uh, you know, I see credits as almost like a bucket of…… of, of a currency, a virtual currency. Uh, you know, you, your company could be X, so it’s a credit X, any credits, right? So you can cre- you can buy those credits and consume those credits. So it’s a f- it’s also a future unit of entitlement. Something that you promise that your customer will be able to buy in the future.

All right. And a token is? A token is a s- a, a production of an LLM, which can be a word or a chunk of a word. Okay. So if I’m- So you can actually buy credits of tokens, right? You could buy credits of tokens. So if I’m a SaaS company selling business applications, I’m most likely gonna be thinking in terms of credits. And, uh, it’s an abstraction layer that’s- Yeah…. likely more customer-friendly than token, which is more of a machine-to-machine language there. I like what you said, I mean, to your point. Yes, it’s friendly, but also credits can actually be very daunting for buyers, right? So- Mm-hmm…. uh, that’s why, you know, going back to your question, I built this triage tool, which are, you know, four simple questions which helps you decide if you need credits or not. Okay. So what are those four questions, right? Number one- Yeah…. uh, is there highly variable

cost, uh, when, to serve customers? Well, the answer is yes, obviously, for most AI companies. I don’t think I need to, to be too sure. So this one, check. Second, uh, do you have a very diverse set of things you produce, right? So do you produce a lot of different things? Which goes to the third question, for those different things you produce, do they have a, a very strong variance in terms of individual value? And last, can customer actually commit against their consumption? And usually, when you have all these factors reunited, then there’s a very strong likelihood that you’re a good candidate for credits. Mm-hmm. All right. So simple thing, not everyone is going to need credits, but there are- Yep…. credit models in the wild. Most likely, we’re becoming more familiar with them than we have in the past. Yep. And so

make sure you need them. And then once you need them, there are some design questions that you might go through. So again, these are all tools that are available to everyone who’s on this webinar, so we don’t need to go too deep into the decision process. But what is interesting about the consequences of these design decisions is that they all put some pressure on the finance organization. Would you agree? 100%. And to your point, Amy, you have to ask yourself, you know, maybe a subscription would do the trick. Maybe a usage-based model or a committed model with overage can do the trick, and that’s what the triage is about. So if you actually need credits, then, you know, you’re gonna have to ask some very complex questions to yourself. For example, what are credits, right? So is the credit an activity, an output,

an outcome, if you have a list? And can it be several of the above? Uh, then, you know, are credits one-to-one, or are they weighted with different values? Uh, are… Can you use them across the portfolio, or are they ring-fenced? Uh, you know, if I look at ServiceNow, for example, I think they have more than 60 rate cards for credits. Some cost one credit for some actions. S- some cost several thousands, right? So, and other questions, such as, is the credit table locked or does it have a floating rate? So imagine, right, a floating rate is good for you as a, as a vendor, because you can actually protect your costs, so your margin. Uh, but if you’re a customer, then you’re actually buying a value of something that may float. And is the… Can the price of credit actually change as well during the contract?

So these are massive questions, and I’m just giving you some examples, right? Uh, and that’s, that’s really, you know, why, uh, we’ll discuss this after, right? We, I had to design these 12, um, dimensions or these 12 attributes of credits. And one last question, which I’m sure is on all lips, which is, what do you do with unused credits, you know? So first off, do I have a monthly burn or yearly burn? Do I have rollovers? Uh, is there a cap on the rollover? Uh, or do they stay forever? Well, for CFOs, forever isn’t, usually not a good response. But what’s interesting, and I p- I would stop here, uh, when I just look at this, th- th- the statistics that you’ll find again in the article, uh, I could find, uh, very interesting that 46% of companies right now have implemented the ability to top up when you have no credits. But if

you see the other way around, that means that 54% of companies do not enable their clients to buy more credits when they desperately need credits, which is interesting. And something like less than 20% of companies actually do let their customers have rollovers of credits. So we’re still in this, in the Stone Age of, of credits, and things are gonna move pretty dramatically. There’s, uh, they’re really important. If anybody on the, on the webinar here has had experience with LLMs within their own organization- Mm-hmm…. or people who are adopting them and runs out of credits, and can you imagine having a more than a few hour lag (laughs) between the time- Yep…. that you’ve run out of credits and the time that you need them. Um, and imagine that for, uh, uh, you know, any business application in the future, there’s going to be a very, very low tolerance, uh, for that.

And one thing I wanna editorialize as well, uh, you know, we, you and I, talk with a lot of organizations as they’re going through these pricing decisions, and not always is the finance organization involved as deeply as product and engineering. Particularly as people are moving really fast, product and engineering are building these tools really, really quickly. And I do wanna caution people. If you’re on this webinar, you likely have a finance orientation or, you know, you, you’re, you’re understanding the implications for finance, but I absolutely expect to see companies bumping up into a lot of challenges from a financial operations perspective if they have not involved the finance organization in these types of design decisions, monetization decisions and questions

because of the implications. And so really the importance of involving your controller from day one, probably preaching a little bit to the choir here, uh, from this, for, for this particular audience. But I would say, if you’re having a difficult time within your organization getting product or engineering to kind of be thinking about this, what I would suggest doing is making a cheat sheet for that organization of, you know, here are the types of questions that we need to be asking ourselves as we’re designing these models. We know that, uh, what we set up today is likely gonna change in the future. We know that the pipeline of products is moving faster and faster than ever before. And, and, and, and so making sure that, uh, the implications are really well-understood

throughout your organization so that you’re not setting yourself up for, uh, for challenges later on. So three things that you can do right now, um, from a metrics perspective, and I’m just gonna summarize real quickly so that we can get into Agentic Finance. Um, picking that metric, again, all of the, the, the, the tools for doing that are available in Michael’s Compass framework, as is the, uh, the determination matrix for whether or not you need credits, and, uh, the questions that you need to ask yourself from a finance perspective once you determine that you do. And then finally, uh, I alluded to it a little bit, but making sure that there is a clear owner for this. Someone who really understands the overall implications, uh, that governs pricing, po- packaging,

portfolio decisions…. as a single discipline, cross-functional discipline, wh- whether you have a pricing committee, where that sits, you’ve got to make sure that you have representations across, uh, businesses. That’s what we’re seeing a lot of the names that we mentioned in this section. That’s how they are running this. And, and so there’s both a sort of a strategic as well as an operational solve for these moni- monetization questions that are really important. So congratulations, after you’ve run this framework, let’s say you’ve, you’ve, you’ve gone through a sprint and you’ve solved for monetization. But you’ve also made your finance team’s job exponentially harder, right? So Michael, uh, we’ve already given examples of

companies that have offers that co-exist with one another and, and, you know, when you go through that matrix, if you’ve got a high amount of, of diverse capabilities, you likely have different things coexisting. Each one of these is adding complexity, right? More invoice line items, more decisions on how to allocate, more amendments. Maybe you’ve got credits that are rolling over that you have to keep an eye on and understand how those are run through your system. And then speed, right? Everything is moving more quickly. It used to be that you’d change pricing, like when I started, like e- every few years. Then it became maybe once a year. And now

what do you… what would you say, Michael, as far as the iteration of speaking? When you have to. When you have to, which would be (laughs) probably every six months. So to your point, Amy, this, this slide should be useful for monetization for now. (laughs). For now. For now, yeah, in, in, in parentheses. So then we’ve got this operational complexity in an already stretched and lean finance organization, so you’re, you’re l- likely already running lean. Um, we’re not seeing as many new accountants coming into the pipeline. And those that are, you know, we’re all being faced with new circumstances that we haven’t been faced with in the past, so we’re learning as we go, uh, about how would we… how all we do this. So this is leading up to the next topic in this presentation. How do you

solve for this? Well, of course, with AI. It’s, uh, you… AI is, is creating all these opportunities and challenges and, and it is a way out for solving some of the organizational complexity. And so I’m gonna really simplify down. We go to… We do a lot of dinners and have a lot of conversations with finance executives who are looking to understand what are other companies doing, how are they sorting that out. So I want to start a little bit of, like where you would start with this, what is the maturity curve look like, and, uh, and then we’ll be wrapping up this presentation. So one of the things that we found really helpful for the companies that we work with for our customers is that they are prioritizing where to start with AI across

five different dimensions. And I will say, you know, we are talking about finance here. We’re talking about an incredibly mission critical, have-to-get-it-right audit, et cetera, et cetera, uh, and so it’s really, really important that, that we all keep that in mind. You, you know, we… I am, I know you are, uh, on the webinar, that this, this can’t be something that we’re experimenting with without that human oversight to begin with. So what are companies doing early on? Well, they’re looking at high volume. Um, how often does the process occur? This is something that happens pretty frequently. How much manual effort is involved? Is it rules based? Do you have those rules documented?

How much human judgment is involved in the process? How much risk is involved? Is it customer facing? Is there revenue tied risk associated with it? And then looking at the data. W- when I talk with a lot of companies about how they’re thinking about AI, data always comes up, either, “We’re not ready,” or, “We’ve got all of this data.” And I think people are getting pretty savvy. We’re all experimenting with things that we can point AI at, where it’s going to really help us parse through large datasets and surface information and surface patterns in a good way. And so that focus on the high volume pattern-based work, uh, some of the early examples that we’re seeing traction with, things like collections and cash app, billing ops, revenue recognition

and close. Of course, you all know that entire quote to cash processes, uh, what are those processes your finance team runs every day? These are all of the different areas where AI can absolutely show up as a quote to cash provider. Zora has lots and lots of, uh, initiatives across quote to cash, where we’re helping customers in various degrees, uh, using AI. And so early benchmarks of what good looks like. Again, thinking about that prioritization framework, what are some, uh, some workflows or capabilities that meet those five conditions, where we’re seeing people have early wins? Cash application match rates, any type of t- uh, thing when you are trying to find

an anomaly or line up information or data, you can have that human oversight, that human check. But that’s something that’s absolutely, uh, a, a… Seeing an AI-enabled, uh, huge, huge jump there from proactivity perspective, QA cycles on bill runs, change analysis, subscription change analysis, going from a manual process that used to take 30 minutes to now you’re getting a batch of information to review in less than five minutes. Disclosure prep cycles going from eight hours down to 30 minutes. So-The other really important thing to see and to understand is that as companies are starting to, to break out individual workflows and processes and apply AI to these,

these tools get smarter over time. And so as they handle more transactions, more contracts, get involved in aspects of more close cycles, they start to learn and understand your unique patterns of your business, and that creates the opportunity for surfacing anomalies, right? For anticipating what questions might come up from the auditors, and from getting… You know, they’re, they’re getting measurably better and we’re seeing that in our own lives with the tools that we’re working with every day. So from maturity perspective, this is McKinsey’s framework, and it actually follows… It was when I was looking at your three-by-three matrix, uh, Michael, it’s- it’s- it’s pretty similar, right? First is the ask, right? AI informs, but the human does the job.

And that’s where we see companies really starting, right? It’s gonna answer a question, it’s gonna look at datasets and detect anomalies, and it’s gonna come back to you to actually make a decision or to do something. The next stage is the AI actually executing, but you reviewing and approving. So, it might be an individual task along a process, maybe prioritize collections. You know, hear the things to, to look at right away this week. Maybe it’s contract change analysis, but the human still comes in and approves. And then the piece that’s a little bit more science fiction probably for most of us is the actual orchestration, the AI owning the workflow, a human overseeing, and that might be an audit package as an example

that might be flagging exceptions. I think if, you know, if I were presenting this three months or four months ago, this orchestrate would be like, “That’s never gonna happen,” but I’m talking to more and more people that are thinking, well, if we look at this process by process, and if we start to see the types of improvements that we’re experiencing in our own lives with some of, you know, even just the basic LLMs that we’re interacting with, that are getting to know us and our patterns, there’s still a lot of slop out there that we’re sorting through, but we can see a path to eventually these things getting good enough to start to do discrete activities and processes on their own with us overseeing them. So, those are the types of… Uh, that’s the progression. Again, focusing on those five areas

of priorities, what could you do in the next quarter or so to think that through, building the map against those stages, look through that quote-to-cash lifecycle, assess, you know, where do you have those high volume workflows, where are you spending a lot of manual time, where do you have really rules and guardrails already set up, where do you have a lot of data and prioritize that way. And then starting with those two to three high impact, but low-risk automations, where you’re spending a lot of time, where you can do that eight hours to 30 minutes compression, that can really help you demonstrate a win and build some confidence in the approach. And then thinking end-to-end, right? If you start to build processes and you get that repeatability

and the AI starts to get smarter and smarter within those processes, where there are opportunities to join things together. So, your a- ask starts to get closer to an act, which starts to get closer to an orchestrate. So, it’s a really important reason why we talk about the, uh, the- the- the improvement that we see over time and why I think finance and AI and finance is really going to help us within this function transform strategically and operationally, how we’re able to… Going back to what Michael talked about in monetizing AI. So, if you just add AI to finance, for example, you know, do a standalone

metering there, or a pricing configuration, you can see some early wins today and probably a lot of you on the phone, on the webinar rather, are experiencing some of that. But when you think about the context that exists within your end-to-end order to cash or end-to-end financial operations, contract history, amendment patterns, uh, credits, right? Credit lifecycle and breakage credit- credit, uh, you know, the ability to, um, be more flexible to be able to deliver new credits to customers, for customers to roll over credits, SSP allocation. All of this context creates an incredible rich repository for Agentic Finance, and that’s where we expect companies to really, really start to see the value

of AI and of finance, and enable the flywheel. So, I just wanna close with one thing. As we were putting this presentation together, uh, Michael and I have spent a lot of time on the monetization by design and in seeing what’s happening in AI monetization, and we’ve also spent a lot of time on Agentic Finance and being at a company like Zora and having a lot of conversations with finance and seeing how our customers and how leaders in the market are starting to gain some operational efficiencies. But when you start to put these things on the same slide and map them together, you realize there’s some real clear dependencies and accelerants that exist. So, in order to be adaptive in your monetization, in order to have that agility to

solve monetization for now and to be ready for what’s next, you need to have Agentic Finance Operations, A- AI Agents that are embedded in the system of record, so that they can keep up with all of the agility, right? That- that they can add those guardrails where the human processes might have broken down or not been fast enough, where they have the audit trail, where they can explain what is happening with this consumption model or that credit model and, and, and really have that ability to operate within existing controls. Um, think of that Agentic Finance of what a lot of companies used to do manually with RevOps or DealDesk, they’re gonna need to automate now with Agentic Finance end-to-end in order to make this

work. And then when those, when- when you have AI and finance, and when you have that context layer, you can achieve the flywheel effect… you start to have compounding intelligence, right? That full customer context brought into every single interaction. And the ability for the, the CFO, the CIO, for FP&A to look across the entire quote to cash cycle. And then when you have that, you have better data to inform your monetization model. Because your monetization model, you know, we are having a lot of conversations right now with companies that are trying to understand, for every customer, for every feature, what is my cost, right? ‘Cause we talked about AI breaking that, that margin model. What is my cost f- to, to deliver this capability to this customer

or this feature to this customer at this period of time? How much am I spending on the LLM? What does that big picture look like? And how does that aform- inform my monetization strategy? And then how do I then adapt my monetization strategy? And then how does that impact agentic finance operations? So, we really see that, that as these companies start to de- to build capabilities in each of these areas, enabling this flywheel that will allow you to solve for monetization, to solve for the squeeze that we see within our finance organizations and within the market, to have that compunting, and compounding intelligence that comes from that rich contextual layer, and then finally to use those insights to help drive better,

fairer, more equitable customer outcome-focused, uh, monetization models. So, next steps. So as promised throughout, there are, are assets that we’ll deliver to you that will help you run your own models, your own organization, through the compass framework, through the revenue architecture report, which will help you understand how finance is adapting to this moment, and how you can, uh, you know, look at the, the metrics and measurements within your financial operations to help you meet the, the moment where it is. And then finally, Zora AI for Finance. So with that, I wanna thank you so much. Thank you, Michael. Uh, I wanna turn it over to Doug. But before I do,

Michael, any last things that you wanna make sure we underline and put an exclamation point on? Absolutely. It’s gonna be very simple. What I’ve heard from you, Amy, right, there’s an explosion of models, right? Usage-based, subscription, credit, all living within one company, if not within one contract. And on top of that, rapid, you know, revenue model situations. And that, in my view, creates a compounding time bomb on finance, you know, which can lead to restatements, which can lead to, you know, things that we don’t like to happen. So, to your point, I think, you know, this, this ledger-grade, AI power, code to cache capability I think is gonna be of the essence to actually power this AI monetization revolution. Amazing. All right. Thanks, Michael. And over to you, Doug.

Awesome. Thanks, Michael and Amy. I thought that was fantastic. Really appreciate all the insights there. And so, I will quickly transition here now over to our demo. So, I think that really the key thing that I heard from, from Amy and Michael, you heard it a few times, is, is really agility. People wanna move faster, whether they’re feeling the squeeze, margin pressure, eroding, uh, licensing costs, uh, as seats drop, um, or they’re just looking to experiment. You know, Michael had mentioned maybe before, it was adjusting pricing every year or every three years. Now people are really iterating much more frequently. We see maybe every six months now, people are adjusting. So, people wanna try out new models, try things quickly,

learn quickly, and be able to adjust on the fly. The other key piece here is, is really the cross functional element here. Uh, it’s really finance, sales teams, IT and engineering, that all have to work together quickly to be able to solve these challenges if you wanna operationalize and scale any of these AI monetization strategies. And so, that’s really where I wanna focus the demo today, is really on, how do we solve for that end-to-end, from pricing and packaging, to the go to market motion, to the visibility that underpins all of that, so that you can truly move quickly, launching new offerings, learning, and adjusting on the fly? And so, what I wanna start out with is actually the Zora monetization catalog.

This is really a key building block when it comes to deploying your new pricing and packaging. I wanna walk through a few sample plans here. Uh, Michael had mentioned kind of outcome, output, activity, all these different kind of, uh, usage models that are out there, what we’re seeing in the market, some of the most common ones. And then, we’ll transition to the go to market. And then, like I said, we’ll wrap up with visibility and kinda how you can learn and see at a per customer level, as well as a, a per cost level, as you look to try out and experiment with these new offerings. And so, I’ll start with really the most simple option that we see, uh, a couple of the most simple ones, which is your pay-go or pay-as-you-go plans, and your, your AI overage plan here.

This is really companies that are looking to tweak and adjust maybe their standard plan, protect their margin by add- adding an overage cost that, you know, maybe as you come with… And let me go in and actually edit one of my plans here. So, think about this as you have your base subscription. That base subscription could be tiered, where you have, you know, maybe a, a pre-built number of units that have a, a set cost. You have scaling costs. As people consume more, buy more, maybe that, that price per unit drops as you have those larger enterprise deals. That base subscription can come with a, a default quantity. So, you know, maybe you’re buying a, a thousand credits that comes with your base plan.

As I over-consume that……I can incur my, my overage charge. That’s kind of the, the base model, one of the more simple options, uh, that we see. You’d mentioned, Amy, Michael, you know, rollover, really a key part of that as well. So, you know, maybe customers to start are just saying, “Hey, we don’t wanna roll over as part of our, our initial go-to-market strategy.” Uh, I see a lot of times this is refined as people come back through sales testing and you start to get customer feedback. Maybe we come back and we wanna adjust and start to add some rollover periods, a little bit more grace periods as customers. Or maybe under consuming and don’t wanna have that breakage at the end of their contract. Um, you know, really the key thing that I wanna highlight

here in the product catalog, one, you know, very easy to configure different pricing models, whether it’s, uh, a base plan, in this case, tiered, flat fee per unit, volume-based pricing. Lot of different out of the box options that we can experiment with. I think this is one area where I see people move very quickly is, is not having to do custom build projects to account for, you know, whatever custom or, or specific pricing model that you have in mind. We can also have finance controls really built upfront. I think I saw in the Q&A, you know, people… I think the big fear in finance is that either you slow things down from a sales perspective or at the end of the day,

after you launch this new pricing and packaging, you’re left with a bunch of manual work at the end of the month, trying to reconcile all these different usage events. So having those finance controls upfront, at the point of configuration, whether or not you wanna do revenue recognition rules, allocation. Uh, Amy had mentioned SSP. All of these different revenue recognition, uh, rules that typically are an afterthought, bringing them forward in the pricing and packaging discussion helps finance be more strategic, I think as part of that pricing and packaging discussion, but also helps prevent that manual work at the end of the month where, hey, we launched some great pricing and packaging, the customers love it, and now finance is leftover trying to figure out what happens at the end of the month.

The other kind of more complex model that I’m seeing, and I think, uh, Mike would mention it with ServiceNow, I see this across the board with a lot of other customers as well, is you have this base subscription, comes with a set of pooled credits, and then you have that rate card. I think Michael had mentioned ServiceNow has 60 different rate card options. I did a couple sample ones here just to show some examples. But this is definitely something I’m seeing very commonly across my customer base where, you know, your base subscription comes with a set number of credits, and then I have all these different activities, outcomes, outputs, all these different things that I can do that maybe have a different drawdown rate. So I’ve heard the analogy of the state fair analogy, where a customer is paying a… $10,000 a month. They get access to a

pool of, of credits or, or tokens, kind of a, a currency that they can use, and all of these actions, we’re gonna have ungated access to all of our different features, whether it’s sites or workflows, builds, whatever might happen and be a part of my, my core platform. And all of those might have different varying drawdown rates against that commitment. And so this is definitely probably one of the more common models I’ve seen, where you have this variable drawdown rate against that base subscription. And once again, having that ability to have those revenue recognition rules, finance rules upfront helps as well as, you know, the ability to have that tax built into this as well. We’re really limiting all those different manual work that might happen as we spin

up and experiment with new options. Something else that I wanna just talk a little bit about as well is the, the concept of entitlements. And something that I see being a, a slowdown when it comes to deploying new pricing and packaging is you spin up a new price, you know, a new model that you wanna do, and then IT has to figure out how does that fit into our home-built entitlement system. You know, we have all these different features that we’re launching, and now we have to map what this new bundle, what this new package is, and how does that correlate to provisioning on the backend? So bringing those entitlements and that raw usage data through a meter into the pricing and packaging is also another way where, you know, potentially a, a manual bottleneck, something that IT is gonna have to take on as

a custom project. We start eliminating that work and, and really accelerate the ability to, to launch and go live with a new package. And so you can see here a number of different features. This is our feature dashboard. Uh, these can be either metered or unmetered features. So in this case of sites, I have a metric that’s associated with that token count. We can set different policies associated with those features. So, you know, if I wanna hard block or cap a feature as somebody over consumes, or maybe I wanna allow or, or warn an account or a customer as they start to near those thresholds, you know, for what they’ve committed. And so having that ability to bring in the raw usage event,

associate it with a feature, not only does this help with cost modeling, which we’ll kind of wrap up with, with visualization, it also helps, you know, really limit the amount of IT work, engineering work, that’s needed to ultimately launch and go live with your go-to-market strategy with the new pricing and packaging. So, you know, once again, very easy to set up either metered, unmetered. These can be Boolean or, or Enum, uh, access to different models or true/false, uh, features as well. So you can start to build out your feature catalog and associate those as you saw in the product catalog to different offerings. So let’s talk a little bit about the go-to-market. So built out kind of some sample products here…. quickly

show that again. I got my full enterprise plan. I got some PayGo plans, maybe an AI overage plan. Uh, how do we start to get sales involved in the conversation? And how do we start to get that feedback from the customer? Well, the great thing is, you know, really getting that automated, that connection between the product catalog and your top of funnel. Whether that’s your CPQ tool, in this case I’m showing ZoroCPQ, uh, so that the sales team can start doing those sales tests, start to test and, and really do those pressure tests of the new pricing and packaging against maybe new deals coming in the pipeline, as well as having that connection to your e-commerce portal as well. So, this can be deployed through a Zoro experience page, your custom website,

but bringing that product catalog and really taking it to the top of the funnel so you can easily expose it either through PLG motions, sales assisted, whatever it might be. So you have that easy, quick connection from the actual core capability there, the product catalog, all the way up to selling and, and going live, uh, with that new pricing and packaging. And so, you know, really that connection, I’ll say this is one area where you see slowdowns as well. You know, if you kind of have that as an afterthought, like let’s say I, I build out my pricing and packaging and then I want to think, “How do I want to connect this to my custom CPQ?” Having to recreate that product catalog in a, a different bespoke system, or how do I connect that to m- my e-commerce portal,

once again becomes really a barrier to moving quickly, testing, eliciting that customer feedback, learning and adjusting. Like I’d mentioned, adjusting things like rollover periods, number of credits that may be included in offering as you start to get that real life feedback. So, the other piece of the go-to market is, it’s not just, as we say, like you have your, your base product, you sell it, it works great. Uh, there’s a lot of real world scenarios that you have to account for. I think Michael had mentioned one of the most shocking statistics I saw is, I think it was 47% of customers today allow for top-ups. Uh, which is, is pretty unusual I would say. So being able to have that top-up

option, roll over, automatic upgrades, having that order orchestration, so as customers start to actually live and breathe with a new product, having that agility to be able to account whether it’s ramped deals, you know, all of these different scenarios, breakage, that actually occurs as part of launching a new AI monetization offering, is incredibly key to actually successfully going live. You can’t think about top-ups, you can’t think about ramps and all this stuff as an afterthought, because once again, you have to go back to the drawing board of your product catalog, rearchitect that, and it becomes really, you know, something that is a bottleneck as you actually launch and go live with a new offering. So accounting for those real world scenarios, having that connection to your e-commerce

portal or your CPQ portal, so that you have that quick, agile launch, getting sales involved, and once again making sure finances aren’t a bottleneck slowing down that sales test. Put it back on sales, get the feedback and be able to adjust and be agile with the new models. So the last piece I wanna talk about is visibility. Definitely not an afterthought. I’ll say there is, um, you know, a few different ways that I like to think about visibility, especially when it comes to operationalizing any of these AI offerings. The first is internal. You know, being able to have a strong control and view on real time trends, and once again we’re consuming that raw usage data so it’s not siloed

across different systems. Being able to see internally where customers are trending, over the limit, near the limit, uh, healthy usage. Having that kind of visibility across the board from your features, your meters, all tied to your new pricing and packaging, is key. There’s also, instead of just the internal controls as well, there’s also the external views as well. And a few different ways that I see customers do this. One, you know, customer dashboards. Once again, this can be a Zoro experience page, this can be something that’s deployed internally. Having that visibility for their direct consumption, whether it’s a build, a workflow, a report, whatever it might be. But giving them that clarity, so

it’s a clear, understandable experience from the customer perspective. It makes sure you don’t have that back and forth with your customer success teams, with the customer creating that friction that can really lead you rolling back any of these new pricing and packaging strategies. Also incorporating ways, easy ways of upgrading plans. I see a lot of customers doing an automated upgrade, uh, talking through or maybe making this part of a, a sales expansion opportunity, so bringing sales involved. Everybody loves an upsell and being able to flag those, those customer accounts that are really prime for a potential upsell or maybe trending to overage, uh, on their consumption. I’ll say the last piece on visibility here,

if I was to go back here into my dashboard, is really around events and notifications. And so the system has a, a lot of out of the box alerts. Probably the most common one that I see is our prepaid balance thresholds, uh, that customers are using. So these are easy to configure, whether it’s a percentage, you know, I see customers maybe at 50%, having an email go out to a customer, 80%. So not just relying on the dashboards and the internal controls surfacing that data internally, but also driving kind of more of that conversation directly with the customer with some out of the box alerts is also key. Now, the last piece I want to talk through. So, this is all current usage, right? We’ve talked about, you know, how customer teams can view and

access that data.Uh, I wanna talk a little bit about also how customers, or how internal teams can start to be a little bit more, uh, gamify the experience, if you will, which is really around projections, uh, simulations, forecasting, uh, and bringing in cost into the equation as well. I know Amy, Michael talked about margin a number of times. You know, this is really key, I think, for making finance a strategic part of, of this function. It’s not just, you know, “Hey, we have this new pricing and packaging. Launch it, set up the revenue recognition rules, and, and let sales run with it.” It’s really about getting people involved and being able to say, “Hey, we have, you know, a potential pricing and packaging that we’re going live with,

uh, a particular feature. I wanna run a simulation and see, you know, what a customer would look like if they were to, you know, over-commit.” Let’s say they, they’re gonna do 100,000 credits, you know, just as an example, just to see. You can start to actually gamify the experience and see what that customer experience is going to be like. You know, we have a hard block in this particular, uh, piece on my, my feature, so I’m gonna see an Access Denied, Limit Exceeded. And so you can start to play with that situation of, you know, what does it look like if, if the situation comes up? Without having to go and run a bunch of usage records, do an upload, or push some fake dummy usage data to us via an API.

It’s really easy to start to experiment with these different models, checking entitlements, and, and seeing kind of, uh, what these different scenarios might play out, and how they might actually impact our different entitlement policies, what may be limits we might set as part of a subscription. So all these different experiments that we can start to run as a finance org, as a team, uh, to start to inform really that pricing and packaging as close to the inception as possible, and this is really just helping you accelerate that iterative process. Now, I don’t wanna steal Catherine’s, uh, thunder too much, but we also have this AI monetization simulator I’m just gonna quickly touch on a little bit. And I think this is really where I see

a lot of this going, is it’s not just about projecting usage, uh, playing with different models, making sure I have the right credits or token limit for a particular plan, uh, checking to see whether it’s per activity, per outcome, per output as a, a unit of measure I wanna drive. It’s also about getting that margin, that safety from, uh, “Hey, I wanna go live with a, an enterprise plan that comes with 100,000 credits.” What does that actually look like from a cost goods perspective? How does that compare to what I’m going to be billing? And how do I make sure I guarantee at the end of the year I hit my margin goals? And once again, having these projections, being able to start to play with some of these models is just another way we see,

you know. This is stuff everybody, I think, is, is very interested in and, and helps elevate the finance work to instead of being an afterthought, being more of a driving force. Uh, Amy had mentioned, you know, somebody needs to own the strategic function that sees really the full breadth of what this touches, and I think this is just another tool in your tool set that can help, uh, as you elevate in those conversations. And I know I am at time, so I will stop my screen share here. And I think I saw a few questions come through on the chat. I don’t know, Catherine, if we have time for Q&A or if I ate up a little more time than (laughs) than expected. Sorry about that. But, uh, happy to pass it over to you Yeah. No worries, no worries. Um, we can go through some of the questions in the chat for the folks who do have time

to stay on. The, um, AI monetization simulator that Doug actually pulled up at the end, that’s, that’s a completely free resource that we’re actually launching, so that’s in your Resources tab. So if you hit the button that says More, and then there’s a button that says Resources underneath that, that’s actually linked there, and we’ll send it to you guys as a follow-up as well for you. Yeah. And Sabrina linked it in the chat too. Um, but we also have Nav here, uh, to help answer some of the questions that we came through in the chat. So, happy to start rolling through some of those. Yeah, I see, Alexandra, a lot of questions here, which is great. So, I think I saw some answers coming back on forth on the metrics, uh, to move away from tokens. Trying to answer some in the chat as well. There’s a question

on entitlement. Is it a custom object or a standard object? So, I think it’s a question from Vivian. Um, entitlement is a core object, which is basically, um, linked features or linked to products, and then when you associate or sell a plan, entitlement becomes associated with plans and then you sell through creating orders. That becomes a grant. I can have token grant of 100,000 tokens or I can have a license grant. Those all are part of the core of Zora. They are not custom objects. I hope that answers it. Thanks, Nav. And then, Sai, I know you asked a question about covering the rev rec for these models. Um, another call to action coming out of this is, you know, I think especially with AI, we’re seeing an acceleration

with POCs and actually how do we connect the dots between how I’m thinking about usage, how my, my finance org, uh, is accounting for this today, and what would that look like in a potential future state as I launch a new offering? Would love to set up some time, Sai, uh, to, to do a follow-up and do something specific to get not just into the details of rev rec, but maybe some of the other questions you might have around, um, some of the models you’re thinking about. Credit calculators. I think, Steven, you were asking about the credit calculators. I went through that a little bit on the simulators. Not sure if there’s any follow-up questions there as well.The concept of credit wallets, as really, as you have that prepaid committed amount

that’s associated with your base subscription, you know, thinking of that as kind of your, your wallet of potential credits, that pool that can be drawn against. Uh, and then being able to do things like calculate, like I showed with the simulator, uh, what it might look like if a, a user was to exceed, uh, their credit consumption. As well as looking at prediction, I think, you know, looking at some of the cost of goods sold margin, uh, pieces that, that Catherine shared in the AI monetization simulator. I think also prediction and forecasting, I think is, uh, a key piece moving forward here as well. Uh, for adjusting so quickly, how do you quickly update en masse 70,000 subscriptions? It’s a great question, Stacy. Um,

I see a lot of… As part of these AI monetization offerings, I see a lot of bulk updates. Uh, there’s a few different ways I see people do this. You know, I think the question, one of the first questions people ask, or that we need to figure out when it comes to these mer- migrations is, am I co-terming this to the end of the agreement and then doing a, a switch to a new plan at the end of the, the life or the end of the renewal? Am I doing this mid-term? Uh, you know, figuring out the timing there, and then, you know, doing basically just a, a bulk order action, where I have, uh, you know, my rate plan swaps or an adjustment that goes through and, and can look at different logic of which subscriptions I want to be able to update. And do those either as a, a contract

nears their renewal or, like I said, mid-term changes. So, uh, work flow is, is probably the most common way I see that done, but there’s, you know, a lot of different ways I’m sure we could a- address that, Stacy. That’s probably the most common one. I think the last question I saw, it’s Steven, “How are sales teams respond to the introduction of credit models?” Um, I will say, this was… It, this can be one of the biggest slowdowns that I see with, with companies actually going live with AI monetizations. You build a new pricing and packaging, you get it in sales’ hands, and then there’s some hesitance about having those conversations. So I think there’s a, a big piece of enablement around making sure sales is comfortable articulating

the story, the, the reasoning, the benefit for the customer as well, ’cause a lot of times this is backed by, you know, you heard Michael and Amy talk about really the, the value tied to these outcomes and outputs. You know, there’s a good story there, and just making sure sales is comfortable with that I think can help address some of the hesitancy. Uh, but I will say, salespeople, they’re, they, they want to just sell how they want to sell. So sometimes there can be, uh, some, some creative friction in that process. But overall, I would say, they tend to respond to what the customers are feeling. So if the customers are resonating with the pricing and packaging, I think sales comes along very quickly as well. Another thing, just on the sales front just to touch on, is that, like, the closer that, like, finance teams are to understanding

the commercial intent of the sale side, the easier that it comes to support some of the things on the back end. Like, really unpacking what the actual goal is you’re trying to encourage. I’ll give a specific example. Um, we’ve heard recently that some companies, like, quite surely don’t want to charge overages, and they just want to- Yeah…. use that as a mechanism to be able to engage customers. Rather than feeling like their customers are penalized because they hit, like, a certain limit or a certain amount, what the sales teams actually want to go and do is use that event of, like, hitting a limit to go in and actually g- engage in conversations, and potentially do an upsell, potentially, like, onboard them into a new, uh, a new contract. Um, so that’s just one example of how finance teams can potentially get closer to the actual commercial intent, just versus just understanding sort of the mechanisms

that need to be in place. I think that was all the Q&A I saw in the chat. I think Michael had answered the other one about metrics that I think, um, I think Tina had asked about. And for any folks who still have lingering questions, we’ll also make sure to review all the chat and make sure we follow up with folks individually as well. Uh, but I thank- thanks for the folks who did hang on and had questions. Um, thanks for all the engagement in the chat as well. Um, Sabrina had linked some upcoming events that we have, and you’ll receive all the resources in the resources tab as well as part of the follow-up for this. So, really appreciate everyone joining today. Um, Doug, thanks for the demo. Amy and Michael,

amazing content at the beginning, and, and thank you folks for joining. Thanks all. Thanks all. Thanks.