Importance of Usage-Based Pricing for AI

Fascinating new data from a16z on how CIOs want to buy AI apps: 1. CIOs now *prefer* usage-based models. So much for usage-based pricing being seen as a barrier to adoption 🤷♂️ My two cents: most AI spend w/ the CIO is on LLMs and compute. Both are usage-based categories where folks have become accustomed to granular pricing and not buying shelfware. With usage-based models, there’s also a connotation that prices will come down over time as vendors compete. Avoid picking a pricing metric that positions your app as a commodity w/o differentiation! 2. Seat-based models aren’t the key to unlocking budget. Only 21% said seat-based pricing was their preferred model. That’s slightly less than hybrid models (23%) even though hybrid models are inherently more complicated with more risk of overages/extra fees. My two cents: buying in a way that reflects the value received is >>> than having perfect budget certainty/predictability. And flexibility matters as folks are finding which AI apps are “experiments” vs ready for “production”. 3. There’s a lot of work to educate CIOs about how to pay for outcomes. 15% said outcome-based models are their preferred way to buy — although that’s still significant considering only *5% of companies* have outcome-based pricing today (based on the 2025 State of B2B Monetization report)! We’re still in the early innings. Among the concerns CIOs raised: (1) lack of clear, measurable outcomes for most products - 47%, (2) unpredictable or unscalabe costs - 36%, (3) difficulty with attribution to a specific tool - 25%. My two cents: I suspect line of business is much more open to outcome-based pricing than the CIO — after all, the CRO and the CMO are measured based on outcomes, too! But if you’re thinking about outcome-based pricing, you need outcome consistency, attribution, measurability and predictability (CAMP). — Cool to see this new data. Very fascinated to see how it changes for next year. #ai #genai #pricing

Per-seat is no longer the atomic unit of software. Consider customer support software Zendesk: companies currently pay per support agent ($115/month/seat), but when AI can handle ticket resolution, the natural pricing metric becomes successful outcomes. If AI can handle a sizable proportion of customer support, companies will need far fewer human support agents, and therefore fewer Zendesk software seats. This forces software companies to fundamentally rethink their pricing models to align with the outcome they deliver rather than the number of humans that access their software. If you are increasing the productivity of labor or usurping it, how should you price this? If every action your customer takes incurs a corresponding cost through an API call, how should you factor that in? How will buyers react to pricing models they’ve not seen before? There’s a lot to consider. However, AI-native companies are leaning into this shift. For instance, Decagon, an AI customer support platform whose AI agents autonomously resolve customer service tickets, offers per-conversation (usage-based) and per-resolution (outcome-based) pricing models to their customers. Both models scale with the amount of work completed (i.e. value delivered) vs. labor (software seats). Read more on Emerging AI Pricing Models in the a16z Enterprise Newsletter with Ivan Makarov and Equals 👇

Salesforce just fired the starting gun on a seismic shift in how we pay for software. At Salesforce #Agentforce, they announced they’re moving away from the traditional per-seat SaaS model to a consumption-based pricing for their AI agents. This is huge. Why? Because it signals the end of paying just to have access to technology. Instead, we’re moving toward paying for outcomes—the actual value delivered. Think about it. In a world where AI agents can perform the job functions of entire departments, does it make sense to charge per seat? Probably not. Here’s what’s changing: - From access to outcomes: Companies will pay for what the AI actually accomplishes. - From subscriptions to value: Pricing adjusts based on usage and results. - From Software-as-a-Service to Agent-as-a-Service: Technology that collaborates with you as a partner This isn’t just a tweak in pricing—it’s a radical upending of commercial models for large SaaS companies. What does this mean for businesses? - Budgeting will evolve: Costs align directly with value received. - ROI becomes clearer: Easier to measure the direct impact of technology investments. - Greater flexibility: Scale usage up or down based on needs without worrying about seat counts. It’s an exciting time, but also a challenging one. Is every SaaS company ready to embrace a model where companies pay directly for the value they receive? At Uniti AI, we’ve been thinking along these lines. We price our AI agents based on the amount of work they do, not on how many seats a company has. I believe this is the future. What do you think? Is the per-seat model on its way out?

I've meant to write my thoughts about this for a few weeks now, better late than never. It's not true that nobody knows how to price AI Tools, as the headline of this article suggested. The issue today is that nobody has the time to properly understand their value, and vendors are trying to price around the problem. The pace and speed at which AI innovations occur set it apart from previous technology cycles. The answer to "What can AI do for my business/function?" changes almost monthly. On a macro level, I think this creates a growing information asymmetry between vendors and customers. Simply put, (AI) vendors currently understand the value potential of AI much better than their customers because they have the talent, time, and motivation to dive deep into the innovation curve. While their customers, for the most part, haven't had the time to figure out what this means for their business. The problem, then, isn't that vendors don't understand how to price AI tools. I believe in most cases—especially considering GOOG or MSFT's talents—AI tools are priced in line with their economic value. The problem is that most of their customers can't leverage these AI tools to their potential yet—not even close. By the time they figure out how to adopt AI for one use case, the technology has moved three steps forward and can enable ten more—all of which a dozen AI vendors are knocking on their door to sell. Looking at this problem through this lens is helpful because this "new" AI pricing problem suddenly looks like an old problem that the SW industry has had to confront since the dawn of time: Innovation (feature) fatigue. When I worked at Microsoft 20 years ago, a frequent complaint I heard from customers was: "I just finished upgrading my company to Office 2003; I'm three versions behind. Why am I paying upgrade maintenance?" At Snowflake, it was more than a little tricky to sell a prospect on the benefits of a unified data cloud when they were running a 20-year on-prem ERP and had a dozen on-prem databases, some running on a desktop under someone's desk. Microsoft overcame its problem (for a time) by leveraging its market power. It essentially strongarmed its customers into accepting a draconian maintenance pricing model(Software Assurance). Snowflake overcame its problem by fully embracing the "land small and expand" strategy with an infinitely flexible and scalable pricing model with usage-based pricing (UBP). Snowflake offers the more applicable lesson for AI vendors today, which is why I am so bullish about UBP for AI. While UBP is, in some sense, an economic necessity (the cost scales with use), the true advantage of Usage Pricing is that it can scale to the smallest use case and provides the flexibility to allow customers to embrace AI however they choose. Start small, embrace experimentation, give the customer time to prove the value of the technology, and then scale up. Usage-based pricing excels at this when done right.