Satya Nadella (CEO of Microsoft) proclaimed that business applications will be replaced by Agentic AI as software evolves into an AI-first era. Software-as-a-Service (SaaS) will make way for Service-as-a-Software models in which entire processes or steps will be fulfilled by autonomous AI agents. Recently, major software vendors have announced initial pricing models for Agentic AI solutions, such as Salesforce Agentforce (per transaction), Microsoft 365 Copilot (included with limits), or ServiceNow and Zendesk AI Agents (per resolution).

While a common pricing model for Agentic AI has yet to emerge, this technology is expected to disrupt the business models of incumbents and emerging software vendors alike.

Defining your Agentic AI pricing model early on is not just a financial decision—it’s a strategic one that impacts product positioning, customer adoption, and long-term profitability: 

1. Pricing Defines Your Market Position and Business Model

→ Your pricing model signals your product’s value

→ It influences how you compete and what you compete on

 2. Revenue Models Dictate Growth and Investment Needs

→  Traditional SaaS models rely on predictable, per-user subscriptions

→  Agentic AI introduces new complexities in usage and measurement

→  If you don’t plan early, costs may outpace revenue, making scalability difficult

3. Pricing Impacts Product Development and Cost Structure 

→  AI agents generate unpredictable costs (e.g. LLM processing, storage, compute). 

→  If pricing isn’t aligned with how the technology scales, profitability suffers

4. Early Pricing Strategy Helps Win Customers and Investors 

→  Customers need clear pricing to evaluate Return on Investment (ROI)

→  Investors look for revenue predictability

5. Changing Pricing Models Later Can Be Costly and Complex

→ Switching from “per-user” to “outcome-based” later can be disruptive for customers and cause churn

→ Salesforce, ServiceNow, Zendesk, Clay, etc. are already experimenting with AI pricing.

→ Early adopters gain a competitive advantage while others scramble to adjust (think: Salesforce's disruption of the on-premise software market) 

Your Agentic AI pricing model is as important as your product to ensure it scales with your business.

↪ How will you address this shift? ↩

Pricing is a complex topic; even more so when moving from a "per-user" to a "per-outcome" model.

Traditionally, costs have been predictable, scaling linearly with the number of users.

In an Agentic AI model, cost becomes much harder to predict. AI agents...

• ...leverage LLMs to process, understand, and generate natural language, resulting in token counts that can vary significantly.

• ...rely on natural language to communicate with each other and access data stored in vector stores, adding to the token count for the length of conversations and potential rework.

• ...work with or manipulate business objects that vary greatly in type, size, and volume, causing additional costs for storing, processing, and updating.

That's where "Pricing Agentic AI Products" comes in:

↪ This first-of-its-kind course prepares you to participate in the new economy of Agentic AI. ↩

Throughout four 90-minute-long live sessions incl. Q&A, you will gain foundational knowledge to price your Agentic AI products. All live sessions are recorded, so no matter where you are, no matter what time zone you're in, you can learn how to price Agentic AI products.

This course is specifically optimized for Product Managers, Product Marketing leaders, and Founders like you. It is designed to be actionable and will include slide presentations, visual frameworks, live sessions, practical resources, interactive Q&A, and community networking.

↪ Join this cohort to disrupt your niche rather than be disrupted. ↩

Interested in an enterprise license so your whole product management team can take the course? Please reach out directly via LinkedIn: https://linkedin.com/in/andreasmwelsch