# Dodo Digest: Every Pricing Model Gets Tested Eventually > Anthropic's new flagship model is far more expensive than previous generations, and the whole AI industry is shifting toward usage- and outcome-based pricing. It sounds fair until invoices grow every time customers succeed. The real edge is predictability. Plus, we shipped Dodo Payments v1.101.0. - **Author**: Rishabh Goel - **Published**: 2026-06-13 - **Category**: Newsletter - **URL**: https://dodopayments.com/blogs/newsletter-june13 --- ## Every Pricing Model Gets Tested Eventually ### TL;DR - Anthropic's new flagship model is significantly more expensive than previous generations. - Many AI companies are responding with usage-based and outcome-based pricing. - That sounds great initially, but becomes problematic as customers scale. - Businesses want predictable costs, not invoices that grow every time they succeed. - We think the same principle applies to payments and subscriptions. - We shipped Dodo Payments v1.101.0 with subscription payment retries, business proration settings, improved B2B invoicing, and more. ## Hello everyone, I was reading about Anthropic's launch of Fable 5 earlier this week. It's reportedly the most capable public model they've released so far, but what caught my attention wasn't the model itself. It was the pricing. Fable 5 is significantly more expensive than previous generations, and Anthropic is already moving away from the idea that frontier intelligence can simply be bundled into flat subscriptions forever. That's interesting because it highlights a challenge the entire AI industry is starting to face. As models become more powerful, they also become more expensive to run. And eventually, somebody has to pay for that. The question is who. ## The Signal For the last few years, AI companies have mostly competed on capability. Better models, larger context windows, more agentic workflows, smarter reasoning. But now another conversation is starting to emerge. Pricing. As model costs increase, many companies are turning toward usage-based billing, credit systems, and more recently, outcome-based pricing. On the surface, that sounds reasonable. If the AI creates value, charge for the value it creates. It's a simple idea, and it makes for a compelling slide in a pitch deck. But business realities tend to be more complicated than pitch decks. Claude Fable 5 and Mythos 5 launch artwork The challenge with outcome pricing is that it works best when invoices are small. Imagine paying based on tasks completed, tickets resolved, leads generated, or workflows automated. At first, it feels fair. You only pay when something useful happens. But over time, something interesting starts to occur. The more successful your business becomes, the larger the invoice gets. In effect, the vendor takes a small royalty on your success. That's not necessarily a problem when you're small. It becomes a very different conversation when the software bill starts showing up in board meetings. Most businesses don't mind paying for infrastructure. What they usually want is predictability. They want to know what next month's bill looks like. They want costs they can budget around. And they generally don't want pricing models that become more difficult to forecast as the company grows. ## Where Pricing Starts To Matter This is why pricing is one of the most underestimated parts of building a company. Founders spend months improving products, refining onboarding, and acquiring customers. Then pricing gets treated as a simple decision. In reality, pricing shapes trust. Customers don't just evaluate what your product does. They evaluate how they're charged. If pricing feels unpredictable, trust decreases. If invoices are confusing, trust decreases. If taxes appear unexpectedly at checkout, trust decreases. If billing becomes difficult to understand, trust decreases. Most of these problems never generate support tickets. Customers simply leave. And because that friction happens right at the point of conversion, it often gets mistaken for a product problem when it's actually a pricing problem. ## How To Think About This Here's how to approach pricing and billing as a founder: 1. **Optimize for predictability** -- Customers are often more comfortable with predictable pricing than theoretically cheaper pricing that fluctuates every month. 2. **Make billing easy to understand** -- The fewer surprises customers encounter, the easier it is to build long-term trust. 3. **Treat pricing as part of the product** -- Pricing isn't separate from the experience. It is part of the experience. 4. **Reduce friction wherever possible** -- Taxes, subscriptions, upgrades, downgrades, invoices, and renewals should feel straightforward and transparent. 5. **Focus on your product, not payment complexity** -- The less time you spend managing billing infrastructure, the more time you spend building things customers actually care about. ## What We Shipped If predictability is what earns trust, the billing layer is where a lot of that trust is quietly won or lost. That's the lens we've been applying to our own roadmap. So rather than chase surface area, the work we've been doing lately is about removing the small, recurring sources of billing uncertainty: the moments where a renewal silently fails, an upgrade charges the wrong amount, or an invoice doesn't match how a customer actually buys. A few of those changes are now live: - **Subscription Payment Retries** recover failed renewals automatically in the background, so a temporary card decline doesn't quietly turn into churn you have to chase down. - **Business Proration Settings** let you decide upgrade and downgrade behavior once at the business level, so every plan change is charged consistently instead of handled case by case. - **Business Name Collection for B2B Invoices** captures the purchasing entity right at checkout, so invoices line up with what finance teams actually expect to see. None of this is flashy. But predictable billing rarely is. It's the difference between a customer who never thinks twice about their invoice and one who opens a support ticket, or simply stops renewing. Dodo Payments checkout collecting business name and tax ID for a B2B invoice ## Builder Spotlight One company I've enjoyed watching recently is CatDoes. They're building an AI-powered website builder that helps people go from idea to live website incredibly quickly. In a world where more people are shipping products than ever before, tools like this lower the barrier between having an idea and actually putting it online. As their user base grew across different markets, they needed a payment setup that could scale with them without adding operational complexity. That's where Dodo came in. Instead of worrying about taxes, compliance, subscriptions, and global payments, they could stay focused on what they do best: helping builders launch faster. It's always exciting to see builders helping other builders move faster, and we're happy to be powering payments for them as they grow. CatDoes, an AI agent that builds anything you describe ## One Last Thought The most interesting thing about the AI pricing conversation isn't whether outcome pricing works today. It's whether it still works when customers become successful. Because eventually, every pricing model gets tested at scale. And when that happens, businesses tend to gravitate toward the same thing they've always wanted: Predictability. Not because it's exciting. But because it's easier to build around. The companies that win won't just be the ones with the smartest technology. They'll be the ones that make growth easier, pricing clearer, and operations more predictable for the people using it. Also, join our loving [Discord community](https://discord.gg/dodo-payments-1305511580854779984)! Best, **Rishabh Goel** **Co-Founder, Dodo Payments** --- - [More Newsletter articles](https://dodopayments.com/blogs/category/newsletter) - [All articles](https://dodopayments.com/blogs)