Latest / SaaS Business with Fexingo: Software-as-a-Service Companies, ARR, and Recurring Revenue / How SaaS Companies Monetize Usage Without Churn
Transcript
- Lucas: The most dangerous trap in SaaS right now isn't churn, it's misaligned incentives. When you price by the seat, your growth depends on selling more licenses, even if the customer only uses half of them. But when you price by usage, you're betting that their actual workloads will scale faster than your costs. Luna: Exactly. And that shift changes the entire sales motion because you're no longer fighting against budget cycles, you're fighting against the customer's own internal efficiency drives. Lucas: Let's talk about the transition. Most legacy platforms tried to force usage onto a subscription model with rigid overage caps. It felt punitive. But the winners like Datadog and Snowflake didn't just add usage; they made it the core engine of their expansion revenue. Luna: Right, because once a company hits a certain size, the per-seat model becomes a ceiling, not a floor. They hit a point where adding users is impossible, so usage becomes the only path up. Lucas: So here is the concrete detail I want us to drill into today: the concept of 'sticky usage.' When a customer builds their workflows around your API or your data pipeline, they can't leave without rewriting their tech stack. That is different from churn resistance, that is structural lock-in driven by volume. Luna: But does that actually protect the bottom line? Because if they use less, your revenue drops instantly. There's no safety net of a fixed contract. Lucas: That is the trade-off. Fixed contracts give you predictability, but they cap your upside. Usage gives you exponential upside, but introduces volatility. The question for operators in October twenty twenty-six is whether the market rewards volatility or punishes it. Luna: I think investors are starting to reward it, provided the gross margins hold up. If your cloud costs rise faster than your usage charges, you're just a reseller with bad economics. Lucas: Before we get into the margin mechanics, I want to pause for a second. This show is ad-free because listeners like you keep it running. If these conversations are useful for what you're building or running, honestly, if today was worth a coffee to you, that's the link — buy me a coffee dot com slash fexingo. It helps us keep digging into the numbers that matter. Luna: Yeah, it really does make a difference. Especially when we're trying to verify these complex pricing structures against real-world data instead of just reading press releases. Lucas: Thanks for the support. Let's go back to the data. The key metric here isn't just total revenue, it's 'cost to serve' versus 'price per unit.' When Snowflake went public, the market loved the stickiness. But did anyone notice that their infrastructure costs were scaling almost linearly with their revenue? Luna: That's the hidden risk. Everyone talks about the top-line growth, but if your underlying compute costs aren't subsidized by long-term agreements, you're eating your own margin. Lucas: Precisely. So let's look at the hybrid model. Many companies are moving to a base fee plus usage. Why? Because it smooths the volatility. You get a floor from the base fee, which covers your fixed engineering costs, and then you have variable upside. Luna: It feels safer for the buyer too. They know exactly what their minimum bill will be, so they can budget it. It removes the fear of the surprise invoice at the end of the quarter. Lucas: But here is where it gets tricky. The base fee often gets set low to win the deal. Then, as usage scales, the marginal cost of serving that user might exceed the marginal revenue if the pricing tiers aren't calibrated correctly. Luna: So you end up with a customer who loves your product, uses it heavily, but actually makes you less money per dollar of usage than a small customer would? Lucas: Yes. That is the classic scaling error. You assume that more usage automatically means better unit economics. It doesn't, unless you have negotiated favorable rates with your own cloud providers like AWS or Azure. Luna: That brings up the issue of enterprise discounts. Big customers demand deep cuts on usage rates. Does that hurt the overall pool? Lucas: It can. But here is the counter-intuitive part. Large enterprises usually consume more consistently. A small startup might spike usage for three months and then vanish. The large enterprise provides steady, predictable volume that allows you to optimize your own infrastructure utilization. Luna: So you're trading higher margin for lower variance. That sounds like a valid trade for a public company trying to meet quarterly guidance. Lucas: Exactly. Public markets hate variance. They love consistency. So the pressure is on SaaS founders to prove that their usage model is stable, not just explosive. Luna: And that stability comes from integration depth. If your tool is embedded in their daily workflow, they can't just turn it off when budgets get tight. Lucas: Which leads us to the 'frictionless upgrade.' In a seat-based model, upgrading requires procurement approval. In a usage model, it happens automatically as they consume more. The friction is removed, which accelerates expansion revenue. Luna: But it also means you lose visibility into intent. You don't know if they're expanding because they want to, or because they accidentally overspent last month and are now trapped. Lucas: That's a great point. Trapped usage is dangerous. If a customer realizes they've been billed incorrectly or overcharged, they will churn immediately. Trust is the currency in usage billing. Luna: So transparency becomes a product feature. Clear dashboards showing real-time spend, maybe even alerts before they hit a threshold. Lucas: Absolutely. Companies that offer spending caps or predictive analytics are gaining an edge. It's not just about charging for usage, it's about helping the customer manage it. Luna: That shifts the relationship from vendor to partner. You're helping them control their cloud costs, not just inflating them. Lucas: Right. And that narrative matters. When you're talking to CFOs in October twenty twenty-six, they are obsessed with cloud optimization. If you position your platform as a way to reduce waste while enabling growth, you win. Luna: Because otherwise, you're just another line item they're trying to cut. Usage-based pricing makes your value proposition very visible in their ledger. Lucas: It does. You can't hide behind a vague license fee. Every dollar spent is tied to a specific action. That forces product teams to be ruthless about efficiency. Luna: Efficiency in the code, efficiency in the API calls. If your software is slow or redundant, the customer pays more for it. That's a powerful incentive structure. Lucas: It aligns engineering directly with revenue. Your engineers care about latency and throughput because it impacts the bottom line. That's a cultural shift most traditional SaaS companies haven't made yet. Luna: Do you think smaller startups can pull this off? Or is it only viable for well-funded platforms with heavy infrastructure? Lucas: Smaller players can, but they need to be careful about their own cost structure. If they're reselling third-party APIs, their margins will be thin. They need to own the IP or the data layer to make usage billing profitable. Luna: So it's not just a pricing tactic, it's a fundamental architectural decision. You have to build your business around the cost of delivery. Lucas: Yes. And that's why we see so many acquisitions in this space. Larger platforms buy smaller usage-native tools to integrate those efficient engines into their broader ecosystem. Luna: It creates a flywheel. The bigger you get, the cheaper your infrastructure becomes, which lets you offer more competitive usage rates, which attracts more volume. Lucas: A virtuous cycle. But only if you don't lose focus on the customer experience during the scaling phase. The technical debt from rapid growth can kill your margins faster than anything else. Luna: So the lesson is: don't just copy the pricing model. Copy the operational discipline that supports it. Lucas: Exactly. Pricing is easy to change. Operations are hard. If you can't handle the complexity of metered consumption, you'll drown in it. Luna: It's a test of maturity. Can your finance team track granular costs? Can your product team explain value per unit? Can your sales team sell outcomes, not features? Lucas: All of those things have to click. Otherwise, you're just gambling on growth. But when it works, it creates a moat that is incredibly difficult to cross. Luna: Because the customer is already invested in the data, the workflows, and the habits. Leaving means losing all that accumulated value. Lucas: Right. Usage creates inertia. Not the bad kind of stagnation, but the good kind of momentum. You keep going because it's easier than stopping. Luna: And as long as you're delivering value proportional to that usage, the customer is happy. It's a fair exchange. Lucas: Fairness is underrated. In a world of hidden fees and complex contracts, transparent usage billing feels like a breath of fresh air. Luna: It builds trust. And trust is what keeps a customer from looking at the competitor next door. Lucas: So, if you're evaluating a SaaS investment or building a product, ask yourself: does my pricing model encourage growth, or does it discourage it? And can I afford to serve that growth profitably? Luna: Those are the two questions that separate the sustainable winners from the flash-in-the-pan hype cycles. Lucas: One final thought. As we move through the rest of the year, watch which companies announce new usage integrations. That signal tells you where the market is heading next quarter. Luna: Because innovation in pricing usually precedes innovation in product. They figure out how to charge first, then they figure out what to sell. Lucas: Spot on. It's a subtle shift, but it defines the next decade of software economics. Luna: Whether you're a founder or an investor, keeping an eye on those usage curves might be the smartest move you make this month.