Tomasz Tunguz put together a nice, thorough guide to SaaS pricing strategy, with a particular deep dive on usage-based—worth a read!
Pricing is one of the most important (and challenging) decisions for any startup, and this doc covers a lot of ground.
A few important callouts:
1?? Maximization, Penetration, and Skimming is a nice, simple framework—but pricing strategy is more nuanced. The best approach might be segment-specific, and it depends on:
?? What you’re trying to achieve (growth, market share, profitability)
?? How your product creates value for different customers
?? Buyer needs and preferences, shaped by market dynamics
Don't feel beholden to picking either Maximization, Penetration, or Skimming. The answer may not fall neatly into either category.
---
2?? Hybrid Pricing isn’t only a transition—it’s a high-potential approach with many variations. Tomasz does a great job laying out the pros and cons of per-seat vs. usage-based and touches on Hybrid.
Importantly, Hybrid Pricing is not just one combo of seat-based and usage-based pricing. It includes any model that combines usage with a traditional subscription, and there are many ways to structure it.
One great example is a usage-based subscription, which allows you to get the long-term growth benefits of UBP while keeping the ARR and short-term predictability that traditional models offer.
---
3?? The best pricing research combines multiple methods of willingness-to-pay assessment. Tomasz warns about relying too much on absolute price questions, but sometimes they’re exactly the right approach—especially when budgets are fixed.
The best practice is actually to triangulate across different methods, which might include absolute and relative measures of willingness-to-pay, for example:
Willingness-to-pay research (e.g., van Westendorp) to gauge price sensitivity
Value profiling to understand how buyers perceive competitive differentiation
---
4?? AI pricing isn’t just about productivity gains. Value is not always based on productivity, and you can't assume willingness-to-pay scales linearly with productivity gains (e.g., if AI is 3x more productive, price it 3x higher).
Willingness-to-pay is complicated and depends on many factors besides value. And in any case, productivity only yields concrete value if it results in a cost reduction or if the time saved is reinvested in profitable activities—and that isn't always true.
AI pricing is still evolving, and we’ll likely see multiple models emerge. But the principles for pricing AI are the same as for pricing anything else:
?? Understand value and how it scales
?? Understand your buyers
?? Think about your objectives
These will form your guiding principles for building your model.
Pricing strategy is never one-size-fits-all. It evolves with the product, market, and customer needs. Great discussion—worth checking out!
#Pricing #SaaS #Monetization #GenAI
