AI vs Seat-Based Pricing: What SaaS Companies Should Do

AI is challenging seat-based pricing because it lets customers do the same work with fewer people, and every person they no longer need is a seat they stop paying for at renewal. SaaS companies that depend on seats should reduce that dependency now, by adding platform, usage, add-on and service revenue over the next few years.

Seat-based pricing is in decline. SaaS companies that do not adapt will follow.

The numbers tell the story. In Kyle Poyar's 2025 survey of more than 240 software companies, seat-based pricing fell from 21% to 15% of respondents in 12 months, while hybrid pricing rose from 27% to 41% (Growth Unhinged, June 2025). The 2026 edition shows the shift continuing, with 37% now running hybrid pricing, up from 25% a year earlier (Growth Unhinged, May 2026).

The market sees it too. By mid-February 2026, ServiceNow shares had fallen 40% in six months and HubSpot 69% over the past year, according to FactSet data reported by CNBC. Investors are pricing in the risk because they understand what is happening to the seat.

The AI efficiency trap

Every SaaS vendor promises to "make your team more efficient with AI."

Even if half of those claims are exaggerated, the result is the same. Companies need fewer people to do the same work, and the market is not growing fast enough to compensate.

The data backs this up. TrueUp counted 783 layoff events at tech companies in 2025, affecting 245,953 people, or 674 people per day (TrueUp). Commercial teams are planning the same way. In our RevOps in 2026 survey of 123 RevOps practitioners, 37% said they are hiring less because of AI and only 9% are hiring more. Among companies with a board-level AI mandate, the hiring-less rate rises to 42%.

Gartner puts the long-term exposure at up to $234 billion of enterprise application spending by 2030, about 20% of enterprise SaaS spend. In the words of its analyst, agentic AI "breaks the link between user growth and revenue growth for many enterprise software vendors" (Gartner, July 2026).

The churn cascade

It plays out in three steps. AI makes teams more productive, companies cut headcount, and seats get downgraded.

Why would a company keep 50 seats when 30 people can now do the same work?

We see the pressure in renewals. In our client work, buyers trim licences to the roles that actually use the premium features, ask for cheaper or view-only access for everyone else, and push back on bundles where only some users need the full product. Vendors respond with offers that add seats cheaply now and raise the bill at the next renewal. The trend accelerates as AI improves. Your annual contracts only delay the reckoning when customers are paying for seats they no longer need.

The largest vendors are already moving. On Microsoft's April 2026 earnings call, Satya Nadella said customers are shifting "from the traditional seat model to seats plus consumption", and that nearly 60% of Microsoft's customer service customers already buy usage-based credits (Microsoft).

What to do about it

Aim for less than 60% of your revenue to come from seats within three years.

Six pricing levers can replace seat dependency.

1. Platform fees charge for access to the platform itself, independent of user count.
2. Add-on modules let you monetise specific capabilities as separate packages.
3. Consumption-based pricing charges for credits, API calls or tokens, so the price aligns with actual usage. Our guide to AI pricing models compares seat, usage, credit and outcome-based pricing in detail.
4. Take-rate pricing captures a percentage of the value or transactions flowing through your platform.
5. Integration fees are charged for connecting to other systems and data flows.
6. Professional services let humans do manual work, because not everything needs automation.

Most vendors are not throwing seats away. They are adding meters on top. Bain analysed more than 30 SaaS vendors introducing generative AI and found that about 65% layer an AI meter, by usage or feature access, on top of seat-based pricing, while roughly 35% simply raised seat prices (Bain, October 2025).

Margins are part of the reason. In the 2026 Growth Unhinged survey, the median target gross margin for AI products was about 50%, and only 12% of companies aim for SaaS-like margins of 80% or more. A seat price that ignores the cost of heavy AI usage leaks margin on your most active customers.

Two principles from our Five Dots RevOps knowledge base help here. First, charge on a value metric, the unit that grows with the value a customer receives. Per-user pricing is rarely that unit, and it makes buyers limit who gets access, which throttles the adoption you need. Second, AI products are badly served by seats when one user drives large processing costs and another barely logs in. We saw this with a software company whose AI processing costs grew with the size of each customer project. Per-user pricing hurt its margins on the largest accounts, so it started to price by project size instead. If you do not yet know the right metric, a flat early tier that removes seat friction and collects usage data is a sensible first step. Learn first, meter later.

Moving pricing away from seats touches packaging, quoting, billing and forecasting at once. If you want help modelling it, book a call with our go-to-market strategy team. Our pricing work covers value-based and AI pricing. We model usage, value and willingness to pay, then ship pricing that grows with your customers.

When outcome-based pricing fits

Outcome-based pricing sounds appealing, but does it work? Revenue unpredictability makes investors nervous. Your CFO will question it and your board will want to understand the variance. The whole point of SaaS was recurring, predictable revenue, and many business models are not suitable for outcome-based pricing.

It can work when four conditions hold.

• The outcome is discrete and countable inside your product, such as a resolved support conversation or a qualified lead delivered.
• You control most of what drives the outcome, so the price does not depend on the customer's own process or data quality.
• The customer agrees on the definition up front, including what counts as a failure or a partial result.
• Volume is high enough to smooth the variance, so revenue stays forecastable.

HubSpot is an example of the fit. In April 2026 it moved its Breeze Customer Agent and Prospecting Agent to outcome-based pricing, at $0.50 per resolved conversation and $1 per lead recommended for outreach (HubSpot). Most vendors are more cautious. Among companies introducing a hybrid AI meter, only about 10% use outcome-based meters, against about 35% metering effort and about 55% metering output (Bain, August 2026).

If your outcome fails any of the four tests, use a usage or output meter instead and keep outcomes as a reporting metric.

The timeline

You have five years. Change is slow.

• In years 1-2, introduce hybrid pricing models.
• In year 3, less than 60% of revenue comes from seats.
• In year 5, you reach full pricing diversification.

The good news is that you have time. Some forecasts are more aggressive. IDC predicts that "by 2028, pure seat-based pricing will be obsolete" as AI agents replace manual, repetitive tasks (IDC, October 2025). Either way, ignoring this risk means you probably will not survive the next five years in your current shape.

Your next steps

Your GTM leadership team needs to meet with product to decide which changes secure your future.

1. Audit your revenue mix. How much depends on seat count today, by segment and by product?
2. Identify two or three alternative pricing levers you can introduce within 12 months.
3. Align GTM, product and finance on the transition roadmap.

Plan the RevOps side as well. In usage-based models, revenue starts when the customer uses the product, not when the contract is signed. Our Five Dots knowledge base recommends measuring activation and early use separately from churn, and rewarding sellers on realised usage rather than signature alone, so they stay involved after the deal closes.

The clock is ticking, and the market will not wait for you to figure this out. If you want a partner for the pricing work, book a call and we will start from your current revenue mix.

FAQ

Is seat-based pricing dead?

No, but it is shrinking as the only pricing model. Most vendors keep seats and add a usage, credit or feature meter on top, and large companies above $150M ARR still rely on per-seat pricing more than smaller ones.

Why does AI hurt seat-based pricing?

AI lets customers do the same work with fewer people. Fewer people means fewer seats at renewal, so seat-based revenue falls even when the customer stays and gets more value from the product.

What should replace seat-based pricing?

A hybrid model usually works best, combining a platform fee for access with a value metric such as usage, credits, transactions or add-on modules. The right metric grows with the value the customer receives, not with headcount.

What is a hybrid pricing model?

A hybrid pricing model combines a fixed subscription element, such as seats or a platform fee, with a variable element, such as usage, credits or outcomes. It keeps part of revenue predictable while letting the rest grow with consumption.

Should we move to outcome-based pricing?

Only if the outcome is discrete, measurable inside your product, largely under your control and agreed with the customer up front. Otherwise, meter usage or output and report outcomes alongside the bill.