For most of the last decade, pricing strategy and growth lived in different worlds. Pricing was the upstream function, deliberate, analytical, updated once a year. Growth was the downstream function, fast, experimental, always running tests. They had the same goal but almost never shared a workflow.
That's changing. And the pricing teams that are pulling ahead have already figured out why.
How Pricing Used to Work
The old model made sense for traditional SaaS. A pricing team would do their research, set the tiers, define the limits, and hand everything off to marketing and sales. Then they'd revisit it in 12 to 18 months.
It worked because pricing was mostly invisible once a customer was inside the product. You signed the contract, you paid, and you didn't think about pricing again until renewal. Growth teams handled everything that happened in between: onboarding, activation, getting people to upgrade. Pricing set the rules. Growth played within them.
AI products changed that entirely.
What's Different Now
In AI products, pricing is no longer something that happens before or after the product experience. It happens inside it, in real time, while customers are trying to get something done.
When a user watches a credit counter tick toward zero mid-task, they're not thinking about a contract they signed. They're deciding, right now, whether this product is worth paying more for. That moment is a pricing moment. It's also a growth moment. And the best pricing teams have stopped treating those as separate things.
The questions that come out of that moment, how to display usage, when to surface the upgrade option, what language to use, how to frame the value of paying more, aren't UX details being delegated to a designer. In the companies getting this right, they're sitting in the pricing team's lap.
Where the Shift Is Showing Up
The product is the pricing page now
In traditional SaaS, the pricing page did the work of explaining tiers, communicating value, and setting expectations. In AI products, that job lives inside the product itself.
A usage counter, a limit warning, an upgrade prompt: these are pricing communications happening at moments of real engagement, when the customer is actively experiencing the product's value. The best pricing teams are treating them that way, designing the communication strategy, not just the tier structure.
Iteration cycles look a lot more like sprints
Usage-based AI models change constantly. Customer behavior shifts. LLM costs move. A new feature ships and average consumption looks completely different overnight. A competitor reprices.
The pricing teams keeping up aren't waiting for the annual review.
Pricing teams are watching signals, running experiments on limit thresholds and upgrade prompt timing and credit framing, and updating on a cadence that matches how fast things actually move.
The experiment queue that used to belong entirely to growth now has pricing decisions sitting in it.
The upgrade moment is a pricing problem, not just a growth problem
In AI-native products, most upgrades happen without a sales rep. A user hits a limit at a critical moment, the product surfaces an upgrade path, and they convert or they don't. Growth often owns the execution. But everything that determines whether it works, what the next tier is, what it costs, whether the value difference lands in that moment, that's pricing.
The teams doing this well aren't handing that off. They're tracking conversion rates and funnel drop-off alongside the revenue metrics they've always watched. They're in the room when usage limits are set and upgrade flows are designed, not reviewing the output after the fact.
What It Adds Up To
The companies pulling ahead on this aren't just collaborating better across functions.
Pricing has genuinely taken on a new operating mode: faster, more experimental, more attuned to what's happening in the product on a week-to-week basis.
The strategic work is still there. Value metrics, price point architecture, tier design: none of that went away. But layered on top of it is something that looks a lot more like growth, monitoring, testing, iterating, and treating every customer touchpoint with usage as a live signal worth acting on.
Additional Reading
If the idea of faster pricing iteration resonates, How I Use AI to Accelerate Pricing Iterations is a practical companion to this post. Where this one makes the case for why pricing needs to move at growth speed, that one gets into the mechanics of how AI tooling can actually make it possible.
⚡Want a Fast AI Pricing Diagnostic?
I share pricing diagnostics in a 15 to 20 minute intro call.