Seats vs. Work Performed: Our Take on Fair AI Pricing

In the world of B2B SaaS, particularly for sales tools, the pricing model has been predictable for years: you pay per user, per month. It’s simple, it’s familiar, and for many software categories, it works.

But for a new generation of AI-native platforms, the seat-based model is fundamentally broken. It creates a misalignment between the value a customer receives and the price they pay.

At Narratic AI, we've made a conscious decision to move away from this traditional model. Our pricing is usage-based, built around a unit of value we call "credits." We believe this is not only fairer but also the only sustainable way to build a true intelligence partnership with our customers. This article explains why.

#The Problem with Seat-Based Pricing in the Age of AI

The core issue with a seat-based model for an AI-first platform is that it incentivizes the wrong things for both the vendor and the customer.

For the vendor, a fixed price per seat creates a natural incentive to minimize the most expensive part of their service: the AI computation. The less processing they run under the hood for each user, the higher their margin. This can lead to taking shortcuts—analyzing only surface-level data, limiting the depth of analysis, or avoiding complex queries—to protect profitability. It encourages building a Toyota with a loud custom exhaust pipe, not the high-performance engine required for real intelligence.

For the customer, a seat-based model creates barriers to value. If every user who wants to view insights costs another $150/month, the information inevitably gets siloed. The CEO might not get a license, or the marketing team is locked out from seeing sales insights. This disincentivizes the widespread adoption that would maximize the value of the intelligence across the entire organization.

#Our Philosophy: You Pay for Work Performed, Not for Access

We believe you should pay for the value you receive, and in our world, value is created when our AI performs analytical work on your data. This is why our pricing is based on "credits."

One of the most powerful—and computationally intensive—actions in our platform is the "Learn" process. When you push this button, our AI analyzes your entire historical dataset in your CRM, including every associated call, email, and note. It's an immense amount of data processing that builds the foundational data model of your unique business. This process is expensive, but it's worth it: It allows us to deliver precise, tailored insights that match or surpass human-level intuition.

A fixed seat price simply cannot account for the vast difference in this computational cost between a company with 10,000 historical interactions and one with 1,000,000. A usage-based model ensures you only pay for the analysis you actually need and use.

#Building a "Lamborghini": No Shortcuts on Insight

Our commitment is to get the best possible information out of your data. We don't just look at the transcripts directly related to a deal; we analyze information about every person and company connected to that deal, even if they aren't explicitly mapped. We do this not to ramp up your bill, but because we know that the most critical insights are often found in these indirect connections. We don't take shortcuts. We are building a high-performance intelligence engine, and our pricing model reflects the cost of running that engine at full power for your benefit.

Internally, the cost of this analysis is driven by tokens—the basic units of text processed by Large Language Models like those from OpenAI, Google and Anthropic. More data and more complex analysis require more tokens. Our "credit" system is a simplified way for you to manage this consumption.

#Aligning Incentives: Our Path to a Fair Model

A usage-based model introduces a new question: what stops us from running inefficient analyses just to consume more credits?

We could say, "trust us, bro." But our real incentive is different. We believe the best way to maximize customer lifetime value is not by inflating a single bill, but by delivering so much value that you can't imagine operating without our platform. Our success depends on your success.

Furthermore, we have a strong economic incentive to be efficient with our own token usage. The more efficient we are, the more competitive we can make our credit pricing. As an economist would say, price is determined at the margin; lowering our costs allows us to serve a much larger audience, which is our ultimate goal. We don't want to argue with you over token consumption; we want the value you get to be a multiple of what you pay us, period.

#A Peek Into The Future: A Hybrid Approach

We understand that a pure usage-based model can feel like a gamble for companies used to predictable seat-based costs. Not knowing the final bill is a valid concern, and it's a conversation we are always happy to have transparently. Our pricing model shouldn't be a blocker; it should incentivize the right behavior on both ends.

As we evolve, we will likely introduce plans that blend the best of both worlds. We can envision a future with different types of "seats" that have different capabilities. For example, a "Leader" seat might have the ability to trigger a full, computationally expensive analysis across the entire CRM, while a "Rep" seat might be focused on consuming the resulting insights and next actions.

For now, we believe a transparent, usage-based model is the most honest and effective way to start. It allows us to deliver the full power of our "Lamborghini" engine to you, with the cost directly tied to the work it performs to make your revenue engine more predictable and powerful.