Pricing AI products seems complicated, but the truth is—it’s not. The fundamentals of pricing haven’t changed, even if AI itself is transformative. Whether you’re pricing a powerful platform or a targeted solution, the same pricing principles apply. Let’s explore how to approach AI pricing strategically.
Pricing Platforms
The first question to answer is: Are you delivering a platform or a solution? A platform can solve many different problems. A solution is designed to solve a specific set of problems. Excel is a platform. Quickbooks is a solution. LinkedIn is a platform. Sales Navigator is a solution LinkedIn built on top of their platform.
Pricing platforms has always been challenging. When there is competition, competition drives prices toward cost-plus pricing. Even for monopolies, the desire to reach a broad audience puts downward pressure on prices, meaning we miss out on capturing value from those who are willing to pay much more. Notice that you can do your bookkeeping and accounting with Excel, but QuickBooks specializes in it. QuickBooks is much more expensive than Excel.
In the world of AI, ChatGPT, Llama, Perplexity, and Grok are all examples of platforms. You can ask them anything. They can do almost everything. I would also put Microsoft’s Copilot in this category. Contrast these with Salesforce Einstein and Grammarly, designed to solve more specific problems.
In the AI platforms, we often see pricing based on tokens, which is a way of saying prices are based on costs. This is not ideal from a pricing perspective, but as a platform it is challenging to capture more of the value delivered while still achieving mass adoption. We also see a lot of per-user pricing with caps on usage, but the per-user price is typically much lower than the value delivered (at least in my opinion.)
The key takeaway is that pricing a platform is not based on value. Different users get widely varying amounts of value, but they all pay the same price.
Pricing Solutions
Many companies are not creating AI platforms, but rather, they use AI to develop solutions for their customers. We mentioned Einstein and Grammarly. Think about Zendesk and Intercom for tech support. I use Fathom to record, transcribe, and summarize my Zoom meetings. SEON and DataVisor help detect banking fraud. There are thousands of these use cases. How do you price those?
This is where the good news comes in. The same as you would price any solution. How much value does the buyer receive from solving the problem? If they don’t buy from you, what will they buy and how much will it cost them? These two questions drive willingness to pay. They drive how much you should charge.
Once you understand how your buyers get value and their decision process, you need to choose a pricing metric. You could charge per user per month. Although that is the most common in SaaS, it’s usually not what you want to do. After all, AI is about automating more and reducing the number of users. FYI, Grammarly charges by the user and does very well.
That leaves you with a usage-based metric or an outcome-based metric. Usage-based is how often they interact with your product or how often your product does something. Companies that monitor usage of many different features have much more flexibility in their pricing and incredible knowledge for future packaging and pricing. Writersonic uses AI to write marketing copy, and charges by the amount of content they create for you.
Outcome-based pricing is effective because you charge the buyer for the outcomes they desire, the reason they bought your solution. Intercom, an AI agent built to solve tech support problems, charges $0.99 per resolution. This directly aligns with the buyer’s goal: reducing the cost and time of handling support requests. The more effective Intercom is, the more customers are willing to pay.
If pricing your new AI product or feature feels challenging, it’s not that pricing has changed. AI did not suddenly change the rules of economics. It’s your product. You are producing a new product or feature that delivers value differently than your current products. It’s time to rethink your business model.
If you’d like some help, give us a call. We focus exclusively on how buyers perceive value.
Share your comments on the LinkedIn post.
Now, go make an impact!
Tags: AI, pricing, Pricing AI, pricing foundation, pricing skills, pricing value, value
