Seat-Based Pricing Isn't Dead. It Just Lost Default Status

Every pricing hot take this year says seat-based pricing is dying, killed by AI agents that do work without logging in. Our directory says something more interesting: seat-based pricing is still the single most common model in SaaS, and it has never been weaker. Both things are true at once, and the gap between them is where most teams will make their next pricing mistake.

The numbers below come from Outmano’s pricing directory, which tracks 999 SaaS products with parsed tier structures for 879 of them, as of the directory’s last fetch. Nobody arguing about seats on social media is arguing from a sample this size.

What 879 Parsed Pricing Pages Actually Show

The model distribution, as of the directory’s last fetch:

• Per-seat: 32% of products with parsed tiers (285 of 879). Still the plurality.
• Usage-based: 28% (246). Four points behind and closing.
• Flat-rate: 21% (188). The quiet survivor nobody writes think pieces about.
• Hybrid: 13% (118). Seats plus a usage component, and the most important number on this list.
• Custom-only: 6% (50). Contact-sales as the entire model.

Two more figures frame the market: roughly half of tracked products offer a free plan (50%), slightly more offer a free trial (54%), and the median starting price across products with a published price sits at $25 per month.

Read the distribution honestly. “Seats are dying” overstates it: one in three SaaS products still charges by the seat, more than any other single model, and in categories built on human collaboration the share runs higher still. But “seats are fine” understates it just as badly. A model that used to be the unexamined default for B2B software now holds less than a third of the market, and the two models growing at its expense (usage and hybrid) both exist specifically to charge for something other than humans.

The 13% hybrid share is the tell. Companies rarely rip out seat-based pricing in one move; they hedge, keeping the seat floor for predictability and adding a usage meter for upside. Hybrid is what the decline of seats actually looks like on a live pricing page. It is a halfway house, and halfway houses fill up in exactly one direction.

Why Seat-Based Pricing Lost Its Default Status

Three forces, in rough order of importance.

The value metric stopped matching the value. The per-seat model encodes an assumption: value scales with the number of humans using the product. For CRMs and design tools in 2015, true enough. For products where AI does the work, plainly false. An agent that answers 4,000 support tickets is not a seat, and a customer will not pay for it like one. When the assumption breaks, the model that encodes it breaks with it.

Buyers learned to punish seat expansion. A decade of per-seat contracts taught procurement teams exactly where the costs hide. Seat minimums, true-ups, and the annual “you grew, pay more” conversation created a generation of buyers who cap licenses, share logins, and treat seat counts as an adversarial negotiation. Products priced on usage or outcomes sidestep that reflex entirely.

Competition made the switch a weapon. Pricing model changes are positioning moves, a dynamic covered in our SaaS pricing strategy guide. When a challenger enters with usage-based pricing against a seat-based incumbent, every seat-minimum in the incumbent’s contract becomes the challenger’s best sales slide. Enough categories have now watched this movie that incumbents pre-emptively add usage components before a challenger forces them to.

Where Seats Still Win

A third of the market is not inertia alone. Seats survive where the core assumption still holds, and the survivors cluster in predictable places rather than being scattered randomly across categories.

Collaboration products monetize humans coordinating with humans. When the number of people genuinely is the value metric, seats remain the honest model. Linear’s pricing charges per seat because a project tracker’s value really does scale with the number of people in it, a structure examined in our Linear vs Jira teardown. Notion’s model, dissected in the Notion vs Coda teardown, leans on the same logic: the workspace gets more valuable as more of the team is inside it.

Predictability-sensitive buyers often prefer seats. CFOs can budget seats a year out; metered usage bills produce the cloud-spend anxiety that made “bill shock” a category of Reddit post. Some buyers will pay a premium for a number that never surprises them.

Sales-led enterprise deals run on seats because procurement understands them. A seat count is legible in a way that “workflow credits” is not, and legibility closes deals. This is why custom-only enterprise pricing (6% of the directory) so often resolves to per-seat mathematics under the negotiated surface.

How to Decide for Your Own Product

Strip away the discourse and the decision is one question: does your product’s value scale with the number of humans who log in, or with the amount of work that gets done?

If value scales with humans, seats are honest, and the directory says a third of the market agrees with you. Take the free-plan question seriously too: half the market now bundles one, and a paid seat model with no free tier is increasingly the exception rather than the rule.

If value scales with work done, and any AI capability in your product means it increasingly does, then pure seats will progressively underprice your heavy users and overprice your light ones. The market’s revealed answer is the hybrid: keep a seat floor for predictability, meter the work for upside. That 13% hybrid share is early adopters of what will likely become the boring default for AI-era B2B tools.

If you do migrate, the mechanics matter as much as the model. Grandfather existing customers for at least a cycle, announce the meter well before you bill it, and set the seat floor so that the large majority of current customers pay the same or less on day one; the meter should monetize growth, not repackage the install base. Teams that skip these steps convert a pricing improvement into a churn event, and their competitors’ sales decks write themselves.

And in either case, watch what your competitors do, because pricing model changes ripple through categories fast. When a direct competitor restructures from seats to hybrid, your own sales calls change within a quarter, whether you participate or not. The full landscape of model choices is in SaaS pricing models: what live pricing pages actually use.

The One Move Worth Making This Week

Run the value-metric audit on your own pricing page. Write down what you charge for (seats, usage, flat) and, next to it, what your customers actually get value from (humans coordinating, work completed, outcomes delivered). If the two columns match, your model is fine regardless of what the discourse says. If they diverge, sketch the hybrid version: your current seat price as the floor, one usage meter on the dimension where value actually accrues. You do not have to ship it; you have to know what it looks like before a competitor ships theirs.

Outmano tracks pricing pages across 999 SaaS products and flags every model change with AI analysis of what it means for your category. If a competitor swaps seats for usage next quarter, you will know the week it happens: outmano.com.

Frequently Asked Questions

What is seat-based pricing in SaaS?

Seat-based pricing (also called per-seat or per-user pricing) charges a fixed amount for each person with access to the product, typically per month or per year. It is the traditional B2B SaaS model and remains the most common single model today, used by roughly a third of products with parsed pricing in our directory as of its last fetch.

Is seat-based pricing dying?

Declining, not dying. Per-seat remains the most common SaaS pricing model at 32% of the 879 products with parsed tiers in our directory, ahead of usage-based at 28%. But it has lost its position as the unexamined default, and the 13% of products running hybrid seat-plus-usage models shows how teams are migrating away without abandoning seats outright.

What is the difference between seat-based and usage-based pricing?

Seat-based pricing charges for the number of people with access; usage-based pricing charges for consumption of the product itself (API calls, monitored pages, messages sent, work completed). Seats give buyers predictable bills; usage aligns cost with value delivered. The models suit different products: seats fit human-coordination tools, usage fits products where the work, not the login, is the value.

What is hybrid pricing and why is it growing?

Hybrid pricing combines a per-seat base with a usage component, such as a platform fee per user plus metered credits for AI actions. It grows because it hedges: the seat floor preserves revenue predictability while the meter captures upside from heavy usage, including AI-driven work no human seat performs. In our directory, 13% of products with parsed tiers already run hybrid models.

Does AI make seat-based pricing obsolete?

For products where AI performs the core work, largely yes: an agent handling thousands of tasks breaks the assumption that value scales with human logins. For collaboration products where humans remain the unit of value, seats stay honest. Most AI-era products are landing on hybrid models that meter AI work on top of a seat floor rather than abandoning seats entirely.