The rise of AI has brought not only technological advancements but also significant shifts in pricing models. Traditionally, software pricing was built around access—charging for seats, licences, or subscriptions. However, as AI continues to disrupt industries, we’re witnessing a transformation in how companies are monetising their products. Here are some of the most innovative AI pricing models currently shaping the landscape, along with the key trends driving these changes.
Disruptive AI Pricing Models*
- Salesforce: $2 per conversation (Agentforce)
- Intercom: $0.99 per AI resolution (FinAI agent)
- Microsoft: $4 per hour of usage (AI copilot for security)
- Clay: Per credit (a credit = a data point or action)
- Copy.ai: Per workflow credit
- Zapier: Per task automated
- Captions: Per AI video generation credit
- Kittl: High watermark of AI credits per day
- Synthesia: Per minute of video
- Imagen: Per AI photo edit
- DeepL: Per user & editable file translation
- EvenUp: Per demand package generated by AI
*If your company is listed above and something is wrong email support@brightreach.co.uk so it can be fixed.
Key Trends in AI Pricing
1. The Shift to “Units of Work”
One of the most significant trends is the shift from pricing based on access to pricing based on units of work completed. Rather than charging customers for the ability to use the software (through seats or licences), companies are now charging for the work the AI actually performs.
For example, Salesforce’s Agentforce charges $2 per conversation, while Intercom’s FinAI agent charges $0.99 per AI resolution. Similarly, Zendesk and Microsoft charge based on successful resolutions or hours of usage, respectively. This model aligns with how customers perceive value—paying for tangible outcomes rather than access to the technology.
This shift is a massive paradigm change in the SaaS industry, moving from access-based pricing to outcome-driven metrics. It reflects the growing maturity of AI solutions, which are now capable of delivering specific, measurable tasks at scale.
2. The Challenge of True Outcome-Based Pricing
While “units of work” pricing is gaining traction, true outcome-based pricing—charging only when the desired result is achieved—is still relatively rare. This type of pricing sounds like a win-win situation for both providers and customers, but it is far more difficult to implement effectively.
Outcome-based pricing requires a deep understanding of the customer’s business goals, success metrics, and a system for attributing success to the AI solution. Moreover, as soon as companies start charging for success, customers begin to scrutinise the results more closely. This increased scrutiny can complicate the relationship between the provider and the customer, as defining and measuring “success” becomes a nuanced discussion.
For instance, Chargeflow charges 25% per successful chargeback, a model directly tied to outcomes. However, not every AI provider is ready or willing to take on the risk and complexity that comes with this type of pricing. Outcome-based models demand a higher level of accountability, and many organisations are still hesitant to adopt them, preferring more straightforward pricing based on tasks or credits.
Conclusion
The evolution of AI pricing models is fundamentally changing the way companies approach value creation. As AI continues to advance, we’re moving away from traditional access-based models and towards pricing based on units of work—charging for the tasks AI completes or the problems it solves.
However, the full transition to true outcome-based pricing remains a challenge. While this model offers significant potential, it requires careful execution and a deep understanding of the customer’s needs and expectations. For now, the most successful AI pricing strategies are those that balance the predictability of task-based models with the flexibility of customer-centric outcomes.
In this new era, it’s clear: AI isn’t just transforming technology—it’s redefining how businesses think about value and success.
