Per-Seat Is Eroding: Usage Pricing Pressure in Vertical SaaS

The per-seat pricing model that built modern SaaS is under pressure for the first time in fifteen years. The pressure is uneven – some categories will keep per-seat for the foreseeable future, others have already moved – but the trend in vertical SaaS is unmistakable. Buyers are reading more, comparing more, and asking harder questions about whether a seat license matches the value they actually get. Vendors that ignored this through 2024 are now negotiating discounts they did not plan for, and renewals are getting messy.

This is a look at where the per-seat model is eroding fastest, where it is holding firm, and what the alternatives actually look like in vertical software.

Why per-seat won in the first place

Per-seat pricing solved several problems at once. It was easy to forecast for the vendor. It scaled cleanly with the customer’s headcount. It produced a predictable expansion motion as the customer grew. Procurement teams understood it. The finance system supported it. CFOs liked it because the unit was a person, and people are countable. For a decade, the model was almost unquestioned in horizontal SaaS, and vertical software followed the lead because the playbook worked.

The cracks started showing when two things happened in parallel. AI features changed the cost structure of delivering software. A single user with an AI-powered tool can do the work of three, which makes per-seat pricing look expensive on a value basis. At the same time, customers got better at measuring what they actually used, and discovered that a large share of their seats were occasional users at best and ghost accounts at worst. Once those two truths met in the same renewal conversation, the negotiation changed.

Where per-seat is already losing

Vertical SaaS categories where the work product is the unit of value have moved fastest. Legal research platforms have shifted to a hybrid where a base seat license covers light usage and heavy users add metered consumption on top. Construction project management vendors are pricing per project rather than per user, because the customer’s revenue is per project too and the alignment is clean. Healthcare revenue cycle tools have largely moved to per-claim or per-encounter pricing for the same reason.

The pattern that ties these together is straightforward. When the buyer can point to the unit of output that drives their revenue, they will eventually push to align software cost to that unit. Seat-based pricing in those contexts feels like a tax on having employees, not a price for the value delivered.

The change is not always vendor-led. In several categories the move came because a smaller competitor entered with usage-based pricing as a differentiator, won a wave of customers on better unit economics, and forced the incumbent to respond. Once two vendors in a category are publishing usage-based pricing, the rest have to follow or accept that they will be the expensive choice for the high-utilization buyer.

Where per-seat is holding

Not everywhere. The categories where per-seat pricing has held firm – and probably will – share a few traits. The first is when the seat itself is the value, not the output. Communication and collaboration tools fit here. Slack and Teams sell access to the network, not transactions across it. Per-seat is the right unit because the seat is the product.

The second is when usage is hard to attribute to a single user. CRM is the canonical example. A single record can be touched by half a dozen users in a typical sales motion, and trying to invoice the customer per record touch is a nightmare on both sides. Per-seat dodges the attribution problem entirely.

The third is when the buyer prefers the predictability of a flat per-user bill more than they prefer the savings of a usage-based model. Some finance teams will pay a premium for forecastability, especially in tightly budgeted public sector or regulated environments. Vendors who serve those segments can hold per-seat longer than the broader market.

The hybrid models that are actually working

Pure usage-based pricing has been around long enough now – Snowflake, Datadog, Twilio – that the failure modes are well documented. Bill shock kills champions. Sales motions are harder because the customer cannot easily compare quotes. Forecasting is harder for the vendor. So most vertical SaaS companies that have moved off pure per-seat have landed on hybrid models.

The most common shape is a platform fee plus usage. The platform fee covers a defined seat count and a baseline of consumption. Above the baseline, usage is metered at a per-unit rate. Below the baseline, the customer pays for capacity they did not use, which keeps the vendor’s revenue floor intact. This shape preserves predictability for the customer at low volumes while still aligning cost to value at high volumes.

A second hybrid that is gaining ground is tiered seat pricing with usage caps. A power user seat costs more than a viewer seat, and each tier comes with a usage allotment. This is closer to per-seat but acknowledges that not all seats are the same. Field service software and clinical trial management software have both moved in this direction.

The third pattern is success fees layered on top of a seat base. The customer pays a flat seat fee and the vendor takes a small percentage of value delivered – revenue recovered for AR automation tools, savings produced for procurement intelligence platforms. This is the cleanest alignment, but it requires the vendor to actually measure the value, which is harder than it sounds.

What the renegotiation conversations sound like

The renewal calls I have heard described over the last two quarters follow a similar pattern. The customer comes in with a utilization report. Of the 200 seats they bought, fewer than 100 logged in last month. They want to right-size, but they also want to know what they get for keeping the dormant accounts. The vendor wants to keep ARR flat. The deal that closes most often is some flavor of seat reduction paired with a usage commitment.

For vendors that have not built usage metering, this conversation is hard because there is no other knob to turn. For vendors that have, the conversation gets easier because the customer can compress seats while committing to a defensible level of activity. Vendors who built usage measurement and platform fee components before they needed them are running noticeably less stressful renewal cycles.

What this means for go-to-market

The sales motion changes when the price is not just a seat count. The most successful vertical SaaS teams I have seen during this transition built three things in parallel.

They invested heavily in a value calculator that the AE can walk through with a prospect during discovery, mapping the prospect’s process volume to a forecasted bill. Without that, the variable pricing surface scares buyers off.

They retrained customer success to talk about utilization and outcomes, not just adoption. The CS conversation in a usage-based world is about helping the customer drive more value, because more value means more revenue. In a pure per-seat world, the CS conversation tends to be about preventing churn, which is a much narrower brief.

They published pricing in some form, even if it was a band. Concealed pricing in usage models is a deal killer because the buyer cannot reason about future cost. Vendors that published rate cards saw shorter sales cycles, even when they expected the opposite.

What buyers should ask

If you are a buyer evaluating vertical SaaS today, the honest set of questions has shifted. Ask the vendor how a 30 percent change in your usage would change your bill. Ask how their power users are charged compared to occasional users. Ask whether dormant accounts can be paused without losing the discount. Ask what happens at renewal if you have over-bought seats. None of those questions are confrontational in 2026 the way they would have been in 2022. They are now baseline procurement questions, and a vendor that flinches at them is signaling something about how their renewal motion works.

The model is not collapsing. Per-seat will still be the dominant pricing shape for a lot of categories in 2030. But the assumption that vertical SaaS belongs on per-seat by default has clearly broken. The vendors that adapt early are the ones quietly winning the comparison shopping that buyers are doing under the surface, and the vendors that resist are walking into renewal cycles they will not enjoy.