Latest / Startup Stories with Fexingo: Conversations About Founders, Funding, and Building Companies from Zero / How Ramp Won Corporate Spend by Automating Approval Workflows
Transcript
- Lucas: If you've ever worked at a company that still uses spreadsheets to track purchase approvals, you know the pain: an employee wants to buy a $300 software tool, they email their manager, the manager forwards it to finance, finance asks for three quotes, two weeks pass, and by the time the PO is approved, the team has already found a workaround. Luna: And that workaround is usually a personal credit card and an expense report that gets rejected three months later. Lucas: Exactly. And that friction is exactly what Ramp set out to kill. Today we're talking about Ramp — the corporate spend management startup that's now valued at over eight billion dollars. But the angle I want to drill into is not the valuation. It's the single workflow decision that made them sticky: automated approval chains. Luna: Before we get into the mechanics — remind us how Ramp even started. The founder, Eric Glyman, had a prior exit, right? Lucas: He did. Glyman co-founded Paribus, the app that automatically refunds you when a price drops on something you bought. Capital One acquired Paribus in 2016 — reportedly for around twenty million dollars. Glyman stayed at Capital One for a while, and that's where he saw the corporate card market from the inside. Luna: So he saw Amex and the big bank corporate cards — and thought, I can build something better with modern software? Lucas: Yes. He left Capital One in 2019 and started Ramp with a co-founder, Geoffrey Charles. The insight was straightforward: corporate cards are a distribution channel for software. The card itself is a commodity. The real value is in the control layer — the policies, the approvals, the real-time data. Luna: So they launched with a card that has no annual fee, 1.5% cashback, and software that automates expense management. But a lot of players offer that now. What made them different? Lucas: The approval workflow. Most corporate card products — even Brex and Stripe's — still treat approvals as a manual step. You buy something, then you justify it later. Ramp flipped that: you set policies upfront, and the system enforces them in real time. If a purchase doesn't match the policy, it's blocked or escalated automatically. Luna: But isn't that just 'pre-approval'? Every procurement department has that. Lucas: The difference is that Ramp's approval chain is dynamic. Say a startup has a policy: any SaaS purchase under $500 is auto-approved for engineering managers. Over $500, it needs the VP of Eng. Over $5,000, it needs the CFO. Ramp encodes that in the card itself — the payment actually declines if the approver hasn't signed off. The employee never spends money that hasn't been approved. Luna: That's a behavioural lock. You can't ignore the policy because the transaction literally fails. Lucas: Right. And that reduces the number of expense reports by a huge margin. Ramp claims their customers see a 70% reduction in expense report submissions within the first month. Because if you never spent outside policy, there's nothing to reconcile. Luna: Seventy percent — that's a sticky number. So the approval workflow is the hook. But how did they get the first customers? They didn't have a sales team at the beginning, did they? Lucas: They did not. Glyman has said in interviews that for the first year, Ramp had zero salespeople. The growth came from a product-led motion: founders and CFOs would sign up, get a card in the mail, and then tell their friends. They also used a clever onboarding trick: you could import your existing Amex or Brex transactions into Ramp's software, so you could see the value of the approval system without even switching cards. Luna: That's smart — reduce switching cost. But once they had users, how did they make money? The card is free. Lucas: Interchange fees. Every time a merchant processes a Ramp card payment, they pay a fee — typically around 2.5% to 3.5%. The card networks split that with the issuing bank. Ramp gets a cut. But the margin is thin, so they need high transaction volumes and low acquisition costs. That's why product-led growth is so critical for them. Luna: So the business model is: give away the software, make money on the payment volume. That's the classic fintech playbook. But the moat is the approval data — the more companies use Ramp, the better the software gets at flagging suspicious or wasteful spending. Lucas: Exactly. And that's the part that's hard to copy. Ramp has seen thousands of approval chains. They know that a certain SaaS tool is priced at $50 per seat but a competitor offers it for $30. They can proactively suggest a cheaper vendor. Or they can alert the CFO that a department's spend on travel just spiked 40% compared to last quarter. Luna: That's the kind of insight a human controller would take days to uncover. Ramp surfaces it in real time. Lucas: And the data compounds. More customers mean more benchmarks, which mean more value for each customer. That's a network effect that Brex and Stripe don't have in the same way, because their approval workflows are simpler. Luna: But Brex is also building aggressively. They have Brex Empower, which is their spend management platform. Stripe has Stripe Issuing and Stripe Corporate Cards. How does Ramp stay ahead? Lucas: I think the answer is focus. Ramp is exclusively a spend management platform — they don't do banking, loans, or treasury services. That focus lets them iterate faster on the approval workflow itself. For example, they recently launched a feature called 'Ramp for Procurement' that automates purchase orders and vendor onboarding. Brex and Stripe are broader platforms, so their spend management features are often second priority. Luna: So it's a classic innovator's dilemma: the incumbents have bigger platforms but smaller attention on this specific use case. Lucas: Exactly. And Ramp also has a distribution advantage through integrations. They connect to Slack, to NetSuite, to QuickBooks. When a purchase is approved, it automatically creates an accounting entry. That saves finance teams hours per week. Luna: Honestly, this episode is a good example of why we keep the show ad-free. If you're a finance professional or a founder and this kind of deep dive helps you think about your own tools, that's the whole point. If it was worth a coffee to you, that's the link — buy me a coffee dot com slash fexingo. Lucas: Yeah, listener support is what keeps these conversations independent. No sponsors to please, just the most interesting angle we can find. Luna: Alright, back to Ramp. So the approval workflow is the core, but they also have a virtual card product that lets you create one-time-use cards for specific subscriptions. That's another layer of control. Lucas: Exactly. You can generate a virtual card with a $50 monthly limit that auto-cancels after a year. So you never have to chase down a former employee's SaaS subscription again. Ramp estimates that companies save an average of five percent on SaaS spend just by eliminating zombie subscriptions. Luna: Five percent. For a hundred-person company spending half a million on software, that's $25,000 a year in savings — just from turning off unused tools. Lucas: And Ramp shows you exactly which ones are unused. They have an integration with your single sign-on provider — like Okta or Google Workspace — so they can see which employees have logged into which tools in the past 90 days. If someone hasn't logged in, Ramp flags it as a candidate for cancellation. Luna: That's the data moat in action. They're not just processing payments; they're analyzing usage patterns. Lucas: And that's why their net revenue retention is so high. Once a customer is in Ramp, the value expands as they add more cards, more policies, more integrations. Companies start with a handful of employees and eventually put all their spend through Ramp. Luna: I want to press on the competitive dynamics. We mentioned Brex and Stripe. There's also Divvy, which was acquired by Bill.com. And there are legacy players like Concur and SAP. What's Ramp's biggest threat? Lucas: I'd say the biggest threat is not a competitor — it's the risk that the approval workflow becomes a commodity feature that every bank offers. JPMorgan and Goldman Sachs have their own spend management tools now. If the large banks bundle it with their treasury services for free, Ramp's differentiation erodes. Luna: So they need to keep innovating. What's next for Ramp? Lucas: They're expanding into procurement — the full cycle from vendor selection to contract management. They launched 'Ramp Procurement' earlier this year, which includes a vendor database and automated request for proposal templates. They're also building out ai powered negotiation tools that suggest discounts based on market data. Luna: But that's a different product category — procurement software is a crowded space with Coupa, Procurify, and others. Do they have the distribution to win there? Lucas: The advantage is that they already have the card relationship. If you're using Ramp for payments, adding procurement is a logical upsell. And the procurement team can use the same approval workflows they're already familiar with. So the switching cost is low. Luna: So the strategy seems to be: start with the card, lock in with approvals, expand into adjacent spend categories. It's a land and expand play. Lucas: Exactly. And it's working. Ramp's annualized revenue was reportedly over $300 million as of early 2026, and they're profitable on an EBITDA basis. That's rare for a fintech at their scale. Luna: Profitable. That's a strong signal. So the unit economics work — the interchange revenue covers the software cost and then some. Lucas: Yes. And because their customer acquisition is mostly organic, their customer acquisition cost is low. They've raised a total of about $1.2 billion in venture capital, but they've been disciplined about spending. Luna: What's the one lesson a founder should take from Ramp's story? Lucas: Find the workflow that is so painful that people will change their entire financial infrastructure to fix it. For Ramp, it was the three-week approval cycle for a $300 purchase. They didn't just make the approval faster — they made it automatic, which eliminated the pain entirely. And they used the data from that fix to build a moat. Luna: It's a reminder that the best B2B products don't add a feature — they remove a step. Lucas: Exactly. And I think that's why Ramp has been able to grow so fast without a sales team. The product sells itself because the pain is universal and the solution is immediate. Luna: So for anyone listening who's dealing with slow procurement, the first step might not be to buy Ramp. It might be to ask: what's the approval chain costing us in lost productivity? Ramp makes that calculation visible. Lucas: That's a great takeaway. And if you do decide to try it, the onboarding is self-serve. You can be up and running in an afternoon. Luna: Alright, Lucas. Any final thought? Lucas: Just that the next time you hear about a fintech unicorn, look past the valuation and ask: what specific, boring, financial workflow did they automate? That's usually where the real value lives. Luna: And that's all the time we have today. Thanks for listening.