It’s Time To Get Specific

Rethinking AI Pricing Models

AI is resource-intensive, continuously evolving and gets more expensive as usage grows, which are all characteristics that distinguish it from traditional software. These differences make legacy pricing models, such as licenses, seats or fixed terms problematic because they weren’t built to reflect how AI is consumed or how its value increases over time.

The companies getting pricing right are approaching it with the same rigor they apply to product development. They’re moving from static pricing to dynamic models, dropping consumption-based billing in favor of value-based pricing and focusing on shared outcomes metrics, including efficiency gains, revenue impact and risk reduction.

These are meaningful shifts. It’s no longer enough to describe what a product does; companies have to demonstrate the business results it delivers.

One of the clearest signals of this change is the growing interest in outcome-based pricing. In models like robotics as a service, providers are paid based on completed tasks—for example, per unit produced or per pallet moved. This type of structure is shared risk in action and changes the relationship between provider and customer, tying revenue directly to performance and creating a shared incentive.

Organizational Change And Shared Risk

Changing the pricing model is often the easy part. Aligning the organization around the change can be much harder. Product, finance, sales and customer success teams all need to operate from the common definition of value. In large organizations, especially, that alignment tends to be cultural as much as operational.

These changes are also impacting how leadership teams operate. Technical leaders are no longer just engineers. They are now expected to connect infrastructure decisions with monetization strategies, not just system performance. Finance leaders are moving from oversight roles to becoming co-architects and active partners in shaping pricing and scalable growth models.

I’ve come to find that it is often startups, with less baggage and more flexibility, that get this right from day one. Unlike larger enterprises, they can build value-based pricing directly into their DNA. By contrast, legacy companies often struggle to rewire how products are funded, sold and supported, and can find themselves wrestling with back-end systems that are harder to adapt to new pricing methodologies. But cutting through these challenges can allow for a deeper alignment of products with customer success and greater trust.

The Role Of Customer Empathy

For value-based pricing to be successful, companies need to embed customer empathy into their core operations. This requires understanding the job the customer is asking to be performed, identifying the specific business outcomes they are trying to achieve and gauging the impact of your technology on those metrics.

Now is the time to reframe how you think about pricing. The focus should be less on building features and more on value delivery with measurement built in and communicated with precision.

Start by asking yourself these foundational questions:

• Are we charging based on effort or on outcomes?

• Do we understand what our customers actually value?

• How can we structure our pricing to align with those priorities?

• How can we create value from differentiation?

• Can our pricing model scale with the growth of AI workloads?

• Are we aligned internally across product, finance and sales teams?

• Are we willing to experiment and learn?

In the AI era, your business model has become part of your product, shaping the customer experience, driving adoption and determining whether your innovation gets funded, scaled and is trusted. To lead, organizations must move beyond building great technology and fundamentally redesign their ways of working to meet customers where they are.​

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?