The pricing page is now the hardest page on the site
For fifteen years, B2B SaaS pricing pages converged on one pattern: three or four columns, a feature checklist, a highlighted middle tier, and a per-user-per-month number. It worked because the underlying model was simple — value scaled with people, and people were countable.
That correlation has broken. When part of the work inside the product is done by an agent rather than a person, seat count and value delivered move in opposite directions. A customer who replaces five workflows with automation needs fewer seats while getting more out of the product. Under per-seat pricing, your best outcome is a downgrade.
So the model is changing, and the design problem that follows is genuinely harder than the one it replaces: how do you present a price that is not a single number, to a buyer who has to justify it to a finance team that wants one?
Where pricing is actually heading
The useful framing is a spectrum rather than a replacement. On one end, pure per-seat. On the other, pure outcome — payment per resolved ticket, per qualified lead, per executed campaign. Almost nobody sits at either extreme.
The direction of travel is well documented. Gartner expects at least 40% of enterprise SaaS spend to move to usage, agent, or outcome-based models by 2030, with seat-based revenue share dropping from roughly 21% to 15%. IDC projects that 70% of software vendors will have moved away from pure per-seat pricing by 2028. And Chargebee's subscription data puts hybrid model adoption at 43%, projected to reach 61% by the end of 2026.
Hybrid is winning because it solves both sides. A base platform fee gives the vendor predictable revenue and the buyer a budgetable floor. A variable component captures the value that seats no longer track. The design job is making that two-part structure legible in about eight seconds.
The five things a modern pricing page must do
- State the shape of the model before the numbers. One sentence at the top: "A platform fee, plus usage." Buyers who do not understand the structure never get to the tiers.
- Give a floor. The starting price, unqualified. This is the number that gets you onto shortlists and into AI-generated comparisons.
- Make the variable estimable. An interactive calculator with defaults that match a realistic customer, not an empty field with a slider from zero to infinity.
- Show a ceiling. Spend caps and alerting, stated as a feature. This single element removes most of the procurement objection to usage pricing.
- Define the unit precisely. What is a credit, a run, a resolution, a seat? Ambiguity here is the source of nearly every pricing dispute that follows.
The pricing estimator is the page now
On a per-seat page, the pricing table was the primary object. On a usage or hybrid page, it is the calculator — because the buyer's actual question is not "what does it cost" but "what will it cost us".
Designing one well is mostly about defaults. A calculator that opens at zero and asks the buyer to guess their own monthly volume fails, because they usually do not know it, and a wrong guess produces a number that either alarms them or misleads them. A calculator that opens pre-filled for "a 50-person company processing about 10,000 documents a month" gives them an anchor and lets them adjust from it.
Three details separate the ones that work from the ones that get abandoned: show the annual figure alongside the monthly one, because budgets are annual; show the breakdown between base and variable rather than one blended total, because that is how it will be explained internally; and let the result be copied or emailed, because the person on your pricing page is very often not the person who approves the spend.
What outcome-based pricing demands that usage pricing does not
Charging per result sounds like the cleanest possible alignment, and it is the hardest to put on a page — because it moves the entire conversation onto measurement.
If you charge per resolved support ticket, the page must answer: what counts as resolved, who determines that, what happens when the customer disagrees, and what they pay when the outcome does not happen. If you charge per qualified lead, the definition of qualified is now a commercial term, not a marketing one.
Buyers are not resistant to outcome pricing in principle — it transfers risk to you, which they like. They are resistant to definitions loose enough that their procurement team can imagine being billed for something they do not consider a result. The page that converts is the one that pre-empts that objection in writing rather than deferring it to a call.
Do not hide the price
The temptation with a complex model is to replace the number with a "Contact sales" button and handle it in conversation. It is a mistake, and increasingly an expensive one.
B2B buyers now build shortlists before speaking to anyone, and AI assistants summarising your category cannot cite a price that does not exist on the open web. A competitor who publishes a starting point appears in comparisons you are silently absent from. You never see those losses, which is precisely why the habit persists.
Publishing a floor plus the shape of the model costs you very little negotiating room and buys you presence in every automated comparison of your category. The rest of the page's job — positioning, proof, and objection handling — is the same work we lay out in SaaS website design that converts, and it starts with the same discipline as a homepage: say the thing plainly, first, as covered in homepage messaging.
If you are mid-migration from seats to usage and the pricing page is where it keeps stalling, we would be glad to look at it.