Your pricing metric is not just a billing mechanism. It is a statement of belief about where value lives.
When you charge per seat, you are saying value comes from access. When you charge per transaction, you are saying value comes from usage. When you charge per outcome, you are saying value comes from results. The metric you choose tells buyers what you think they are paying for, whether you intended to send that message or not.
Many companies choose their pricing metric based on what is easy to measure, what competitors do, or what their billing system supports. It may be a convenient metric, but it sends the wrong signal. Buyers feel the misalignment even when they cannot name it.
The right metric is the one that scales with the result the buyer gets. When the buyer gets more value, they pay more. It feels fair because the price and the outcome move together. When the metric tracks something unrelated to value, every invoice feels like a question the buyer cannot quite answer: what am I actually paying for?
Credits Are a Confession
Credits have become the default pricing mechanism for AI products. They feel flexible, buyer-friendly, and easy to explain. But look at what they actually are.
A credit is a company currency. It lets you charge for almost anything and quietly change what it buys whenever you need to, without ever repricing in public. That flexibility is the real appeal. And that flexibility is also the tell.
A stable pricing metric is a commitment. It says: this is what we charge for, and by implication, this is what we believe is valuable about what we do. Credits let a company sidestep that commitment entirely. You can shift what a credit covers, adjust what it buys, and fold ten different capabilities under one currency, all without ever having to say out loud which one of those capabilities is actually worth something and how much.
Credits are a confession that a company has not yet found where its value lives.
That is not always a criticism. AI is moving fast, costs are uncertain, and the problems buyers are solving with AI agents are still taking shape. Genuine uncertainty is a legitimate reason to reach for a flexible mechanism while you figure it out. Salesforce has changed Agentforce’s pricing structure roughly four times in two years, cycling through flat per-conversation fees, flex credits, flex agreements, and pay-per-resolution models. That is not indecision. That is a company still discovering what problem it solves, for which buyers, in which situations.
Credits are the honest answer to “we are still figuring this out.” The problem is when they become the permanent answer.
The Buyer-First Path to a Better Metric
The way out of credits is to find the result first and let the metric follow.
Buyer-First thinking starts with the buyer’s world. What problem are they living? What does their world look like after the problem is gone? What specific result can they point to and quantify? Once you can answer those questions with specificity, the right metric becomes much clearer. It is the unit that best tracks the result the buyer cares about.
The metric you land on matters beyond pricing. It signals to every buyer what you believe is valuable about what you do. A metric aligned with the result the buyer gets builds confidence. A metric that tracks something unrelated to value creates friction, makes the business case harder to build, and makes the price easier to object to.
Find the result. Find the metric that tracks it. Everything downstream gets easier.
My new book, Pricing AI, written with Michael Mansard and Wolfgang Ulaga, goes deep on exactly this.
Share your comments on the LinkedIn post.
Now, go make an impact!
Tags: AI Credits, ai pricing, buyer-first pricing, credit pricing, outcome pricing, pricing, pricing metric, value, value metric, value-based pricing
