Get your free personalized podcast brief

We scan new podcasts and send you the top 5 insights daily.

Frequent price shifts from foundation AI providers like Anthropic and OpenAI create cost unpredictability for enterprise buyers. To provide procurement predictability, martech vendors are transitioning toward engagement- and outcome-based pricing models that align software spend with tangible activations and results rather than seat counts or basic operational efficiencies.

Related Insights

As AI moves from being a simple tool to an autonomous agent, pricing models are evolving. Companies like Sierra, chaired by OpenAI's Brett Taylor, advocate for outcome-based pricing, which charges for delivered results (e.g., a completed report) rather than the underlying token consumption.

Budgeting for AI is difficult because the utility-based, per-token pricing model is not viable or scalable for business departments like marketing and sales. This system is a temporary phase; expect AI providers to shift toward more predictable, outcome-based pricing models as the technology matures.

AI enables a fundamental shift in business models away from selling access (per seat) or usage (per token) towards selling results. For example, customer support AI will be priced per resolved ticket. This outcome-based model will become the standard as AI's capabilities for completing specific, measurable tasks improve.

Wade Foster argues that fixed, seat-based pricing is dying because it doesn't align with the variable nature of AI-driven work. He sees the market splitting into two models: usage-based for commodity-like AI services and outcome-based for enterprise tools that deliver a clear, measurable result (e.g., "price per resolved ticket").

Seat-based and token-based pricing for AI are intermediate steps. The ultimate model is outcome-based, where customers pay for specific value delivered—like a successful investment idea or a completed report. This perfectly aligns vendor cost with customer value, bypassing debates over token consumption ROI.

Bret Taylor of Sierra argues outcome-based pricing (charging for a resolved case) is superior to usage-based pricing (charging for tokens). It aligns vendor and customer interests by tying cost directly to business value, not resource consumption. This forces the vendor to improve product effectiveness, not just optimize for usage.

To combat the unpredictable costs of token-based AI usage, Pega is adopting a value-based pricing model. Instead of charging per token, they charge based on work completed (e.g., per loan funded or service request processed), aligning costs directly with business outcomes and enabling forecasting.

OpenAI is reportedly exploring outcome-based pricing, where customers are charged only if an AI successfully completes a task. This model shifts from a commodity-like 'cost per 1000 tokens' (CPM) to a value-aligned 'cost per successful action' (CPA), better aligning incentives.

In the age of AI, software is shifting from a tool that assists humans to an agent that completes tasks. The pricing model should reflect this. Instead of a subscription for access (a license), charge for the value created when the AI successfully achieves a business outcome.

With AI agents reducing the need for human users, the per-seat SaaS model is becoming obsolete. The future lies in outcome-based pricing, where vendors charge for tangible results like cost savings or tickets resolved. This aligns incentives, forces vendors to prove ROI, and creates a true partnership with customers.

Engagement and Outcome Pricing Protects Enterprise Budgets Against Volatile AI Model Costs | RiffOn