Ramp launches token spend management as AI costs hit 10% of payroll, growing 21x in a year

is TVPN Royalty. We got Eric Lyman from RAMP. He's the co-founder and now the co-CEO. He's in the waiting room and we'll bring him in to the TVPN Ultra Dome.

First time we've talked to him since he's become co-CEO. How is it? How does it feel?

It feels so good to be back. I missed you, too. Um, first time chatting with Tyler directly. I think he's popped in a few times. You've you've chatted with us in every different permutation, the yellow suits, in person in the New York Stock Exchange, all over the place. Um, but uh, how are you settling in into the new title co-CEO?

It feels good. It's uh, what's been so fun and I have worked basically this way together for 15 years.

Yeah. Yeah. I was about to say everyone's like trying to do like hot takes around it and I'm like, have you actually met these guys? They've been co-CEO the entire journey, even going back before RAMP.

It's It's been fun. Like I I uh like like for me, like part of what makes this fun is like for we've known for years like Kareem is the secret weapon of the company. Like he's driving so much of what's going on now. It's like more obvious to people of like we got at least two of us. There's there's actually way more interesting people at the company. But no, we're it's uh we're moving fast, growing faster this year. And um it's just fun. You know, now it's uh he can swap and take some some events off my hands, too, which is great.

Yeah, I love it. Uh well, uh businesses on ramp are moving tokens fast through their systems. It's showing up in their books. You built a website, token-spend.fm. I love the FM. I think that's a very fun uh TLD. Uh but what inspired this? I imagine it was from like direct conversations with your customers. Did this come internal? What were the findings? What were the goals of the project?

You you nailed it. Like o over the last year, the last full months alone, ramp customer spend on tokens has grown by 21 times.

Wow.

21s. Like

is that a gong or is that a

It's a little bit of both.

It It's some of every depends who's making the money or spending the money. And look, by the way, like like the crazy part is

it's not like people are like, "Turn it off." It's like, "No, I actually want to spend more on the right things." And so like,

you know, so we saw this from our customers. We saw this from ourselves. Um, you know, a few years ago, AI spend was a routing error to

la I think in May it hit almost 10% of our payroll spend. The equivalent was on tokens on payroll. And look, there's great things you can say about it. We're launching products faster than ever. Uh we're more efficient than ever. Um uh we're growing faster and yet, you know, you I hate to say there are people at the company who like use Fable uh to like look up the weather. Um

that happened yesterday. We heard about that in other companies. I can't believe I'm teasing. I'm teasing. No, it it's a there's this whole thing where um people know and get in the abstract of models even from a year ago were amazing are great at doing uh tasks and we can be more efficient. uh two accounting teams need to be able to monitor this like it's you know I think of our CFO who would would get a bill for hundreds of thousands of dollars and then the work began of like

they have to allocate some to engineering some to sales and marketing some to you name it and so there's just so many products that weren't built um by the labs and so today's uh launch around token spend management uh is a place where uh any company whether or not you tried rampkin uh can can link up your your your API keys and then you're you're good to go. Um and uh you know we're helping people really within minutes start cutting their spend by several percentage points and so it's been a very fun launch.

Yeah. What was the uh what was the first sort of generative AI application at ramp? Was that the GPT API for understanding receipt data? So this was actually so that was our first ML model for for sure but the generative use case um this I think it was 6 months before chatgpt came out we hopped on GPT3 and we started using this for for uh um go team classifying different uh receipts putting things in the right expense category um things that could be done deterministically fuzzy logic could apply but LLMs were uniquely suited, but at the time that was a rounding air. Then you go forward to May and you get a bill for 1.5 million in a single week. What's the actual process for for untangling what's happening? Are you looking at prompts? Are you just going to Slack and saying, "Hey, you know, you were one of the top 10 drivers of token spend. Can you flesh out a little bit of what what you got done this week?" That type of thing. What is the correct way for an organization to interrogate their spend um maybe qualitatively after they're done with the quantitative side?

Great. Most organizations that are even at a place to be thinking about this,

they're using lots of tools, right? They're spending on open AI models. They're spending on enthropic. They're using Gemini. They might be dabbling in open source. They're using cursor. And so the first step is actually just seeing it. It's being able to link up your keys so you can understand and start to break up basic questions of like what's happening today, not in a month when you go get your bills. Um connecting and tagging that and that's something you can do out of the box through the product.

We come back within minutes to help you understand it. And so you can start to go and see, okay, not just this person normally spends um uh $1,000 a month, but they've already ran through $800 in an hour. And so you can set up notifications. We show you unusual spikes. And so again, think back to the early days of of RAMP and corporate cards. A lot of this was alerting and visibility. It's these types of insights. And you also see things like, you know, we know how the the most efficient companies are running. And so if you aren't uh cashing um uh you are overspending uh some people will leave fast mode on which can be multiple times more expensive and so we'll just highlight that for you and then finally at the end you get to controlling it um you know acting on it and uh you know I think for today there's so much to do on the analytics itself but I do think there's more sophisticated opportunities uh to be had whether that's in in uh small model training in routing and more and we're excited at the whole space. I think there's so much to do.

Okay. Uh sure.

Yeah. I have a question. So I see um the average company uh 59% of their token spend is is on frontier models. How do you how should companies be thinking about allocating between frontier models and open source? Is it almost a thing where like you're in a explore phase, you're using the frontier models, you're you're doing these kind of net new coding tasks, whatever. And then once you find this repetitive thing, you're you're running the same process every day, every week. Is that when you when you allocate it towards open source, maybe you're even doing a fine tune? And how should people be thinking about that kind of stuff?

It's a perfect question. And I I think for companies out there, if you're listening to this and like haven't used these models, like I I I I would say using Frontier models just to get a feel is good. It is surprising the capabilities that models have. And you know, if you if you don't have a multi,000 a month build, like start there, I think is is reasonable. But then you start getting into optimization. And there's two there's a few sets of interesting questions. One um there are cases when using frontier models can in fact be cheaper. Like for example in our own benchmarks um on our software engineering benchmark people think of sonnet as an older model in let's say the anthropic world uh and it is cheaper per call but it needs to think a lot harder uh and call more agents working in collaboration and and actually the smarter models like it's kind of like you met someone smart like they don't need to go and do like long division they can just like do division in their head there are cases where using frontier models

example

Scott Woo calling Scott Woo is much faster. Um, his hourly rate is much higher, but there might be certain things. You know, the Scott Woo per second might be cheaper than you know hiring a team of

certainly the venture capitalist have made. I I'm like I want one Scott Woo instead of a thousand.

This is right. So, so, so, so, so that's part one. But then the the really interesting part is when you get down into benchmarking of what is the nature of work that you're doing

you know you can use these you know trillion parameter models to answer really tough questions but when you start seeing very high production use cases you'll see uh you know if you know all of the input tokens you're getting are around uh customer service you know you'll see companies like Sierra having their own small models around what makes great efficient. And so you they've lobotoi uh you know just the part of the brain that is really useful for those types uh of services to if you can dynamically start to route based on the complexity of the task. You can say a small model just for uh accounting maybe is all we need. We don't need to go ask you know uh a model that can allow us to cure cancer and and do quantum physics and that kind of a thing. And so I in some sense it's on both and you start to get really interesting answers the more that you can benchmark your own business and the more that you have uh high fidelity about the nature of the inputs you're getting the outputs you're seeing and then the efficacy per scent on getting to that uh and there's so much to build around this area. uh talk a little bit about the shape of AI product development, the AI work you're doing, Ramp Labs, sort of the surface area, the spikes there, because there's there's, you know, just using AI to improve the product that is deterministic, right? Better software. Then there's also AI integrations like I mean it's such an anodine feature, but I love the fact that you can open up the RAMP app and just ask a model like how much did we spend on camera equipment last month? and it'll just tell me. And that's amazing. And I'm sure I could like wire it up to some other system. But like I love just having it there. And then there's also like AI research and harness development, all sorts of work that's happening there. And that's can be expensive. But how are you thinking about all the different trade-offs and all the different work within the AILabs umbrella? So on using this, I mean this is just an incredible technology as you know like I I I think about like part part of what like let's talk about like B2B SAS for a second. Um you know if if we must please you know you know come on

I was waiting I was waiting to

Yeah, we're we're finally doing it boys. Here we go. All right. So, uh, the problem why most B2B SAS is awful is companies are complicated. People want different things. And so, the accounts payables clerk wants a view, your, um, your accountant wants a view, the CFO wants a different view, one person wants a button. You want to make it simpler for these people, more advanced for another. And

how do you deal with this? Well, turns out generative interfaces where based off of who you are, how using a product, it can show you the interfaces you need with the views, the graphs that you like to do your work can be intuitive. Uh, and so it's it's very interesting in making um products that are very powerful but feel simple and relevant for the products. And so on one side of it like for forget ramp I just think organizations in in spending time around dynamic and generative interfaces there's I think you can just make better computers you can make better tools for people. Um why do we care as a provider of of of of services? What is so different structurally about spend on models and on tokens and on software is you could just hammer Salesforce all day. You're not going to get like a bigger bill from um

from Salesforce. Sorry, hold on. Uh calling I'll call him back. Um

um uh should Yeah, you're on. I I'll save him for a minute. But um uh to go a bit deeper on it um if you if you start going and using as as folks know in the token maxing era uh lots of tokens like your bill can go from a,000 to 10,000 h 100,000 to a million dollars like very quickly if you start going and it's not like payroll spend that companies are are able to to manage in some way. It's not like normal vendor spend in some sense. It's like an untapped corporate card. uh that you can spin as you go. And by the way, the meter is not running um and you don't see it. And this starts to feel a lot like what we've obsessed over for years of can you help people manage every incremental dollar far better. And so it's pulled us uh deep into answering the question of if you want to know return on investment, we should be thinking about like what is return um what did you buy? Can you understand just the semantics of it? So what was the output to the efficiency as uh there's more types. So it it is uh uh we're obsessing over this like from every floor of the building.

One of the one of the interesting things about ramp is you know everyone who has a ramp card in an organization has you know they can open up and see all their transactions. They can see their budgets their limits the policy that applies to them and they you know they can implicitly understand that you know if they spent $10,000 on their card or $1,000 like did they deliver that much value to the organization? And I'm wondering if you see either ramp or just generally uh businesses need to push more of that data to the youu the end user the the token consumer and how will that instantiate?

I I'll give you an example from like the just card world and we're seeing this already in in in AI spin management. So, um, in the card world, there's this concept of just like out of policy spending. Like, let's say that you

let's say you never got a notice and you spend on Uber Eats because you you forgot to switch over the card on the weekend.

You might keep running it up and never really realize it. And it turns out if you just tell people once, hey, that was out of policy,

you see spend an out of policy spend just drop.

You tell people like, you weren't supposed to do that. They don't do that. Um, you know, people actually want to be good and and do the right thing.

And this is the checking the weather with Fable 5.

Yeah. It's like, I didn't realize I spent $100 to check the weather. Um, and you know, sometimes some subtle UI like this and feedback goes a long way. And so in in in the product, we've already seen um actually just exposing people. here's your AI spend like like like John Tyler I don't know if if you guys know what you spent uh on token spends yesterday but on ramp you can know um and it turns out when you see it you start getting more efficient and so even just the act of exposing it to you um is driving down uh these savings for for people. So you you nailed it. There's so much there.

Yeah. Uh we probably spent a lot on tokens yesterday. We vibe coded a bunch of really really jokey sites that provided a lot of uh a lot of laughs and a belly laugh. We we had Sager and Jetty on the show who who does not like prediction markets and we built him an entire prediction market for his entire life and

he the the belly laugh that he got from that was priceless. It was all play money but I think it was worth it. But I actually haven't seen the token bill so I don't know. It might have been it might have been rough. Um uh I want to talk do you have another couple minutes?

Of course.

Okay. Uh I guess the last question for me um the the ramp econ lab that is a separate lab. I'm very interested in uh the knock-on benefits of Ara Carzian's excellent work uh studying uh the economic impacts all over. I mean the research has gone to the front page of the Financial Times, the Wall Street Journal, so many other places. uh what has that unlocked for you as as CEO in conversations with customers? What have been like the knock-on effects of that that project?

First is just you kind of just understand the world better. Um, it's it's very obvious now cuz it's been this uh I guess to paraphrase Elon like a supersonic tsunami where it's gone from didn't really exist years ago to uh perhaps 1% of the United States's GDP in the next 12 months like it looks quite likely at this point. Um and this data is going so fast like you look at most economic indicators and you you get things a quarter later it's not measured precisely whereas ramp data uh you know we're we're tracking about 1% of all corporate spend in the United States and you can just see it and you can understand it and you can adjust your strategies faster. So I think it actually helps people um better run their business and not get left behind. Um, so I just find it useful for for finance teams um and um technologist people building businesses beyond it. It's just grown awareness. Um, you know, we're competing against some of the

best known brands ever created. And you know, uh, if we're going and and trying to win you over and say, you know, try our tools, um, you should trust us to move money, store money, help you get more from every dollar an hour, and you've never heard of us. It's just a much harder sell to, oh, yeah, I saw RAMP in in the journal on TBPN, um, you know, on the founders podcast or, um, you know, more and more, it it starts unlocking it. And when it shows up in a way where um it's actually already providing value to you before we have a conversation, it it it's uh you can get deeper um much faster and um you know, I think over the long run that that leads to more growth.

That makes a lot of sense. Well, thank you for taking the time to come chat with us. Congrats on the launch. The website token-spend.fm. Go check it out. Optimize your token spend today

with RAM's latest project project. Have a great rest of your Thursday. Have a