"Sell the outcome, not tokens" — selling AI into the enterprise

A credit or a token has actually no inherent value whatsoever. A credit or a token, or if your usage, whatever your action is, whatever your outcome metric is, actually has no inherent value until you associate it to your customer's desired outcome. The more you do, or the more you use, the more you get. So how do we price and package a situation where we just want you to do as much as you can possibly do? It becomes really, really difficult. From a sales perspective in an enterprise conversation, especially when we're talking to a CFO. Subscriptions are relationships, and we need to get back to this concept of like delivering value on an ongoing basis and not on this big 5-year check. It's great to be here. I'm a little bit humbled to be speaking next to all these incredible pricing people because I'm the sales guy. I'm the individual who gets to try to define value of monetizing AI to your customer. So I'll poll the room, like how many of you are interested in selling your AI solutions to the enterprise? Like who wants to sell to the biggest companies or do you want to be PLG and sell to startups? It seems like most people would ideally love to sell to the biggest companies in the world. And so how many of you are already— and you can, if you want to just jump to this slide, it's just a question— but show a raise of hands, like, how many are actively using consumption, usage, or a hybrid model, or want to? And I know that's a little bit more broad question from before, but it sounds like pretty much everybody. And so I'm going to take the approach from an enterprise sales lens. You've heard from a bunch of incredible pricing minds. I'm gonna talk about this all from value creation and what your customer is seeing and what your enterprise CFO cares about. And so there's a little bit of a change in the guard here as it relates to selling AI. And that's really that like pricing and packaging used to be our definition of value. We would have a good, better, best model In the good, you'd have features and things you can do. In the better, you'd have more features and things you can do. In the best, it would be like unlimited. It's awesome. You can do everything. But it was really well-defined, right? These products and packages were telling you what you can do with their product. That's changing. And really, really quickly. Clay's a great example. You can, you can't really use predefined packages anymore. A package is not going to give you the guardrails to say, this is what you do with the product. If you're using Clay, you can do anything really as it relates to data. So how do you define that in a package? It's almost impossible. If you do, you're leaving a bunch of value on the table because somebody who doesn't align with that perception of what you're selling doesn't land on your product and click that trial. And ideally, in theory, I think the reason why everybody loves consumption and Marcos called it out, it's I believe it's his favorite, is that the more you do or the more you use, the more you get. So how do we price and package a situation where we just want you to do as much as you could possibly do? It becomes really, really difficult from a sales perspective in an enterprise conversation, especially when we're talking to a CFO. And so if you jump to the next one here, this is a really pointed question and it's not meant to be a trick question. But it is really meant to kind of double tap on what everyone has said already. And I want you to think about it as you approach your pricing, which is just what is a credit worth? Like, what is the value? And I'm not talking about cost. I don't want to go down the cost-plus rabbit hole and everybody kind of just, you know, fight their way to the bottom. Like, what is a credit actually worth? And so the answer of this is not a fun answer. In my opinion, if you wanna jump to the next one, Joe, it's a credit or a token has actually no inherent value whatsoever. A credit or a token, or if your usage, whatever your action is, whatever your outcome metric is, actually has no inherent value until you associate it to your customer's desired outcome. And so it is outcome-based pricing, right? But I'm really talking about what are you looking to accomplish when you come to our product as a solution. And without that association, credits and tokens mean nothing, at least from an enterprise seller's perspective. I don't wanna, you know. And so value really lives truly in the outcomes that you're talking about from an enterprise sales perspective. Your business of course has value. I don't wanna like, you know, shoot everyone down from the start, but credits gain value only when you know how many you need to be consumed to accomplish that outcome. And that outcome can be very kind of basic things. It can be time savings, it can be revenue gain, it can be accomplishing a high-level operational task within, you know, a Fortune 500 that may not just be like, I saved you $5 or I saved you 5 minutes. It can be pretty complicated. And so we define at Velas use cases as a scoped outcome with 4 key components. It's an actor or actors. Those are the people that are gonna be engaging with you. It's a workflow or workloads, things they're gonna be doing or things that your solution is actually doing for you. The consumption metric or drivers. And so that could be credits, it can be tokens, it could be any type of unit. You can see someone like Databricks just says they're DBUs, they're Databricks units.. It can be really anything. But the fourth component here, sorry, is just the success criteria. Because if we can have all four of those components within what we call like a business use case, pretty standard, you know, business case, we can define why you should spend the money you're spending for the amount of credits we're asking for to accomplish the thing you're asking us to accomplish. And so I think that shift has really started to drive enterprise sales especially within the Fortune 500s. And, you know, we're talking about the larger contracts that we're seeing today. And that's because when you use this concept, concept of a use case, I want to understand when anyone's coming to buy VELUS or when any of my customers are using VELUS to sell their product is what are the use cases that you're coming to solve? The currency of credits is just your way to deliver that use case. So in discovery, when you're talking to these customers, you should be trying to figure out what are the jobs you're trying to do? What are the things we're trying to accomplish? There's been a few incredible talks I've been to recently, and I'm gonna forget the person who said this, but they said, "Sell the work, not the token. Sell the work, not the credit." And it's really stuck with me. What are the high-value use cases that your customer, prospective customer, is coming to you to solve? And that's how we frame an enterprise transaction. Credits is a part of that, absolutely, but it is the currency to deliver the outcome they're looking to buy from you. And then what's really, really great about modeling the use cases is that you now know how many use cases they want to solve. You also know the priority of those use cases. If I'm Clay and someone comes to me and wants to do AI outbound, maybe that's their primary use case, but they also want to automate inbound, or maybe they also want to clean up their CRM. And so, we can start to build these different stacks of use cases, prioritize those use cases, and now we know what's urgent and how we drive the deal forward, because urgency is tied directly to their outcomes, right? And then as a seller, how do you go structure this deal? You're gonna work, of course, with your pricing team and with your corporate strategy team, and you know, you're gonna think about what is the platform, what is the metric, But at the end of the day, everything is tied back to that use case. You're gonna spend a million credits with me, and those million credits are gonna deliver on these three outcomes you've talked about. And those three outcomes, as we can see in the valuation page, and I'll get to ROI in a little bit more, in a second, is that those use cases are much easier to identify ROI. And when we get into negotiations, because We're selling to the enterprise. Procurement's going to come knocking at some point. You're never gonna get away from them. Every bit of negotiation is now not tied to your credits or tied to your pricing model. It's tied to the outcomes they're gonna get. So if they want to spend less with you, they're going to get less with you. And you are able to tie that back to the exact thing that that business was looking to accomplish. Much, much easier to justify spend, to justify the value, of your product because it's tied directly to those outcomes. And so I wanted to keep this relatively short because I think the concept is simple. And I do have a take-home kind of playbook that you can just download. There's a QR code at the end. But I really see use case-based sales as 4 really, really critical benefits. And the first one I already mentioned, but I'll dive a little bit deeper into it. Confidence in ROI. Is incredibly important at the enterprise level. A CFO cares a lot about it, but what they do not like, and I've had this experience many times, is you say, "Hey, it's going to be 11x the spend. You're going to 11x ROI." And they go, "Cool, man. How do you justify that?" And they will not spend the time to actually look at the numbers with you. That's one of the things that I've struggled with in my career is, you you go through a 400-questionnaire to define ROI with every little action that they do, and they just do not wanna understand that level of detail. But what happens when you transition to use cases is now we're looking at a specific task that we're gonna do for the business. Are we gonna automate inbound traffic for people coming to your website and then auto-route them to sales to book meetings? That seems like a great use case. Now, if I look at that and I say, we project that that's gonna increase our inbound booking by 20%, we have a 20% win rate. So if that's 50 deals, I'm gonna close, you know, 10, and we have an ASP of 50K. So now all of a sudden, it's a really easy path to say, whatever we spend on this product, or on this specific use case, has a very clear outcome from an ROI standpoint. And then you can just stack those use cases. And stack that ROI. It becomes way, way, way more tangible, and your CFO's actually going to be able to follow along with you in a very quick pace. And you don't have to just blanket say 11x ROI. Everyone tries to shoot for 11x, I don't know why. Maybe that's why that company named themselves that. And then in addition to ROI, I'll jump to onboarding and customer success because as an enterprise seller, I'm not just gonna close this deal and be done. It's never gonna happen. The most, you know, the strategic and enterprise AEs, they live with their accounts for a really, really long time. And I think you all should as well, because we don't just model use cases for my customer. Modeling a use case is not just to validate the credit spend, although it is a fantastic way to justify credit spend, but it's also a blueprint for how to implement that company. So now I've built a use case of the highest priority thing this customer wants to achieve. I know exactly what the success criteria is of that use case. And now I go to my implementation team and I say, this is the number one thing we have to do. If there's only this one thing that we do with this implementation, this is what we focus on. For a lot of these workflow AI companies that you work with, Clay, NAN, Make, Insert Workato even. One of the biggest problems is you sign up with that solution, you don't know what to do. You don't know what you should do right away to get value from that product. And that's because there's analysis by paralysis. There's too many things you can do. And so what's fantastic about use cases is it just points us a laser at the number one thing the customer wants to accomplish, not anything else about the product. So now your implementation team goes in and says, "Hey, you wanna automate inbound workflows? Fantastic, let's just start there. We won't touch anything else until we execute on that deliverable." What's also fantastic about this is you may model out, and I hope you do, model out as many use cases as you possibly can with your customer. What are all the things that you can do to provide value to them? Get justification from them that those are important, But then prioritize and stack rank all of those use cases. Because you wanna know what they wanna buy today versus what they wanna buy later. And I think a lot of businesses when they're going through a sales cycle focus on selling everything and then it just drags on. Focus on what they wanna do right, right, right, right now. And what's great is when you don't sell a use case, so we sell the top 3 but we modeled 10.

Right?

Now I have 7 use cases that are incredibly clear path to upsell and cross-sell opportunity. Your implementation team went in, delivered the first 2 use cases, and they were incredibly happy. Your very next call is, "Hey, I'm so happy those use cases went well. I know you talked about doing these other things. When are you ready to start that project?" And all of a sudden, your consumption will continue to skyrocket because you're constantly finding new ways to be helpful. And then lastly, I think this is just a byproduct, but it's an incredible byproduct for any of your founders or builders, which is a far more predictable consumption and revenue forecast. And this will continually become more accurate over time because you will start to understand how credits or usage or your outcome-based model is being consumed by use case., which gives you a better understanding of these use cases on average, always execute within this timeframe. So you have a much clearer understanding and you can recalibrate, for some reason that's a hard word for me today, on those forecasts on a consumption-based motion tied to use cases. And so I don't know if there's any other slides, if you just jump in. Yeah, so I wanted to kind of stop there. I think there's a lot of content there and use cases. Are relatively clear, but a lot of people have questions. So I'll open it up, and if you don't have any, I'll answer my own. And then I'll leave you with a QR code you can scan, which is a bit of a playbook on how to get started transitioning to a use-case-based sales motion.

Yeah.

What's your sense of procurement orientations adapting to sort of buying this way? Where like when they do their planning a year before, they're like, oh, we can spend this much money on marketing software. Well, your use case actually might be tabbing into the labor budget potentially where you're not hiring, but procurement doesn't really have oversight into that. It gets a bit more complicated. I guess the marketing team is like, oh, I can buy this and not have to hire 2 content editors. Content marketing folks, but there's only like a $50K budget for content marketing software.

Yeah, it's a great question. So how do you justify that spend when there might not be that budget in procurement?

Well, they're just like unaware, like how's procurement gonna be adapting to sort of like buying AI? Because there's like labor replacement in the mix in addition to Sombra.

Yeah.

Okay, so from an enterprise seller's perspective, and I'll say this with a little bit with a grain of salt, but procurement's job is just to buy the solution that the business needs. And so they may have budget, but it's your job to help the champions within the organization justify the spend. And so they'll find the money where it needs to be found if the business unit has the that really core need for it. And so by breaking everything up into use cases, you can build that story, that business case around that specific need where it's much easier for a procurement team to justify the spend because there's a clear, clear deliverable, clear action, clear ROI. And the conversation then becomes not so much do we buy this product or not, but what use cases do we deliver right now versus what do we need to find budget for potentially, which is the pushback from procurement.

Later.

And so it doesn't necessarily stop you from selling deals, but it may kind of change the way you prioritize the use cases for what's most important now versus what's important later.

Yeah. Since you're talking about enterprise sales, a lot of SaaS companies who are selling enterprises aren't coming to this organically, right? Like this is an evolution from contract-based sales or some other sales model. How do you talk to those companies about talking to their sales teams about that transformation? Dealing with those relationships and how they change with their customers, especially going from, you know, going to a recurring revenue model or as they would think of it, actually having to continue to work for their money, right? Instead of doing a sale every 3 years, they've got to redo retention, try not to let customers churn.

Yeah, so it's actually a little bit of a difficult transition from what I've seen. I think there's a few case studies that we can call out of just like an incredible adoption of this. Snowflake and Databricks are both great examples. Their entire sales motion is based on use cases. And so they're looking at a constant relationship. I think to Marcos's point, like subscriptions are relationships and we need to get back to this concept of like delivering value on an ongoing basis and not on this big 5-year check. You know, I just sold to Amazon and I'm cashing out. I hit President's Club or whatever time. We really need to think about, like, how do we build ourselves into the business and understand them almost better than they understand themselves, so that when they come to us with a problem, our solution is using our product and consuming credits, but it's to solve that problem at the end of the day. So it's hard. Like, I foresee a future where people move off opportunities as an object in Salesforce or their CRM, and they just sell use cases. But I think we're a few years away from that. Yeah.

How do you think about credits across different use cases? Because one use case could be automating an email. Another use case could be a much bigger business case. It seems difficult. If you have one credit, how do you think about, is this actually worth 10 credits? Worth 5 credits? Like, you have different credit tier systems when you're selling to companies or individuals. How do you think about that?

It's going to be unique for every company. And I would probably point you towards the pricing people as the best people to answer that. Although I do think about that a lot for our own business. I look at Clay as a great example. They have different credit spend per action. I think you can have different types of credits. But to Marcos's point, like, try not to have too many because it just gets really confusing. So I tend to, if I was to advise some of the companies that I advise on this topic, I would say like, just understand the value the action is going to give to that customer. And then try and correlate the cost of credits to that value. Now again, credits don't have inherent value until you understand the use case. But if we know that we're going to be able to route 20% more demos, then then we know that it has a pretty high credit cost or credit value associated to it. And so I think you always have to do a little bit of both, which is value and cost, make sure you're not gonna eat into too much action, you're gonna learn over time. But the best thing to do is to start with what is the value you're providing, associate a credit count to that, and then grow from there as you see more information. And you can always pivot, to Marcus's point, I hope that if there's one thing you take away from this as a sales guy is pricing is not set in stone for a year. You should, you should be changing it pretty constantly. Thank you all.