Per-seat pricing became the default in B2B SaaS because it is easy to sell, easy to forecast, and easy to bill. The problem is that it never actually measured value — it measured seats. In products where usage varies widely between customers, which is increasingly the norm now that AI features consume resources unevenly across accounts on the same plan, charging by headcount starts to distort both revenue and the customer’s sense of value.

That is why hybrid models — a base subscription plus a usage component — have been gaining ground. It is not a pricing fad. It is a correction after years of per-seat pricing being applied in contexts where it never fit particularly well.

Why per-seat pricing is losing ground

The commercial case for per-seat pricing was always simplicity: one number, one variable, easy to explain on a sales call. But simplicity for the rep does not always mean alignment with the customer. When one account uses the product heavily with few seats and another uses it lightly with many seats, both end up paying similarly for very different value — and that erodes both expansion and retention over time.

Kyle Poyar’s Growth Unhinged newsletter has been tracking this shift toward hybrid and usage-based models as one of the more structural changes in B2B SaaS monetization now that AI-driven products have made uneven consumption the rule rather than the exception. The point is not that per-seat pricing is disappearing — for many products it is still the right model. It is that it stopped being the automatic default.

What breaks operationally when the pricing model changes

Changing your pricing structure looks like a product or marketing decision, but RevOps absorbs the full operational impact. A few things tend to break first:

  • CRM and opportunity structure: deal value fields built for “seat count times price” do not capture usage components, consumption tiers, or minimum committed spend.
  • Billing and revenue recognition: fixed recurring billing is trivial to automate; hybrid billing with a variable component requires reliable usage metering, tighter close cycles, and more elaborate revenue recognition rules.
  • Sales comp plans: commissioning on closed ACV is simple when ACV is predictable at close. When part of the revenue depends on the customer’s future usage, the rep is being paid on something they no longer fully control — which pushes the comp conversation closer to customer success.
  • Forecasting: a fixed-seat revenue model lets you forecast with high confidence off the pipeline alone. A usage-based component introduces variance the forecast has to learn to absorb, typically with historical data that does not exist yet early in the transition.

How to prepare RevOps before migrating

Most of the pain in a pricing migration does not show up when the new model is announced — it shows up three to six months later, once the team realizes the operation was not ready. Worth addressing beforehand:

  • Have reliable, auditable usage telemetry before pricing on top of it — if consumption data is not trustworthy enough for internal debugging, it should not become an invoice line item.
  • Redesign CRM opportunity fields to separate fixed revenue, estimated variable revenue, and realized variable revenue, instead of a single ACV field.
  • Review the comp plan with sales before launch, not after the first complaint about inconsistent commission.
  • Run the new model in parallel with the old one on a sample of accounts for at least one full billing cycle before rolling it out broadly.
  • Define clearly how existing accounts will transition — grandfathering, timeline, communication — before announcing anything publicly.

A pricing change is not a product project with an operational side effect. It is an operational project that also happens to change the price.

Common mistakes in the transition

The most common mistake is treating the model change as a product announcement and only looping in RevOps after the launch date is already set. The second most common mistake is underestimating how much manual work will be needed in the first months, because billing and commission automations have not been tested in production against real usage data yet.

The third, quieter mistake is migrating the entire customer base at once. Testing the hybrid model on a controlled segment — new customers, or a specific vertical — surfaces data and process problems at a manageable scale, before they show up multiplied across the whole base.

If your company is considering this shift, the real work is not deciding on the new pricing model. It is making sure CRM, billing, commissions, and forecasting can actually operate on it without leaning on a side spreadsheet to hold it together.