Per-resolution pricing charges you when a customer’s problem actually gets solved. Per-seat pricing charges you for every agent license you hold, whether that seat is busy or idle. The short version: per-seat ties your bill to headcount, while per-resolution ties your bill to outcomes. If you want your support spend to track the value you get, per-resolution is usually the cleaner fit. But the right answer depends on your ticket volume, how predictable it is, and how the vendor defines a “resolution.”
This post walks through both models, the costs that hide inside each, and how to choose.
How per-seat pricing works
Per-seat pricing is the old default. You pay a fixed amount per agent, per month. Ten agents, ten seats. Simple to understand, and simple to forecast when your team size is stable.
The trouble is what gets bolted on. The base seat price is rarely the whole bill. Watch for:
- AI add-ons. Many per-seat tools now sell an AI assistant or “copilot” as a separate line item, often priced per resolution or per conversation on top of your seats. So you end up paying twice: once for the human seats, again for the automation that’s supposed to reduce the need for them.
- Channel and feature fees. Live chat, messaging, voice, advanced routing, and reporting are frequently tiered or sold as upgrades.
- Idle capacity. You buy seats for peak season and keep paying for them in the quiet months. Seats don’t flex down on their own.
None of this is hidden in bad faith. It’s just how the model accumulates. The result is a bill that grows with your org chart, not with how much support work you’re getting done.
How per-resolution pricing works
Per-resolution pricing flips the unit. Instead of paying for capacity, you pay for completed outcomes. A customer asks a question, the system answers it correctly, the customer goes away satisfied, you’re charged for that one resolution. Nothing resolved, nothing billed.
This matters for two reasons.
First, it lines up the vendor’s incentives with yours. When a vendor only earns money when a customer is genuinely helped, they’re motivated to actually solve problems, not to inflate seat counts or pad conversation volume. The phrase we use internally is “resolve, don’t deflect”, deflection (bouncing someone to an FAQ so the ticket closes itself) is not a resolution, and it shouldn’t be billed like one.
Second, it makes cost scale with value. Busy month, more resolutions, higher bill, but you got more done. Quiet month, fewer resolutions, lower bill. You’re not paying for empty chairs.
For more on how we think about the unit itself, see our pricing page.
The hidden costs, side by side
Per-seat’s hidden costs are mostly about capacity you don’t use: idle seats, peak-season overprovisioning, and add-ons stacked on the base price.
Per-resolution has a different risk: definition. If “resolution” is defined loosely, your bill can balloon. So before you sign anything, ask exactly how it’s counted.
Questions to ask about “resolution”
- Does a deflection count as a resolution? It shouldn’t. Sending someone a help article and closing the ticket is not the same as solving the problem.
- What happens when the AI can’t help and a human takes over? A fair model charges for the genuine AI resolution and lets the human handoff happen for free.
- Is a single back-and-forth conversation one resolution or several? It should be one.
- Are repeat questions from the same person in a short window double-counted?
- Is there a spend cap so a traffic spike or a bot attack can’t produce a surprise invoice?
Good per-resolution pricing answers these plainly. At Fidiora, human handoffs are free, you set spend caps, and you pay per genuine resolution, no seat fees, no separate AI add-on, no surprise invoices.
When each model makes sense
Per-seat can still be reasonable when your volume is low, flat, and predictable, and your team is small and stable. If you have three agents handling a steady trickle of tickets, a fixed seat bill is easy to reason about.
Per-resolution tends to win when:
- Your volume swings with seasons, launches, or growth.
- A large share of questions are repetitive and answerable from your docs.
- You want spend to track outcomes, not headcount.
- You’re tired of stacking AI add-ons on top of seat licenses.
A quick illustration.
Example: A 40-person SaaS company gets 3,000 support conversations a month, most of them the same handful of “how do I reset this / where’s my invoice / does it integrate with X” questions. On a per-seat plan plus an AI add-on, they pay for agent licenses and for the automation. On a per-resolution plan, the repetitive questions get resolved and billed as resolutions, the rest route to humans with context, and the bill maps to work done. See use cases for more scenarios like this.
Key takeaways
- Per-seat charges for headcount and capacity; per-resolution charges for outcomes.
- Per-seat’s hidden costs are idle seats, peak overprovisioning, and stacked AI/channel add-ons.
- Per-resolution aligns vendor incentives with yours, but only if “resolution” is defined honestly.
- Insist on: deflection not counted, free human handoffs, one conversation = one resolution, and spend caps.
- Per-seat suits low, flat volume; per-resolution suits variable volume and repetitive, doc-answerable questions.
If you’d rather pay for problems solved than chairs filled, take a look at our pricing.
Resolve, don't deflect.
See Fidiora resolve a ticket, capture a lead, and keep the bill predictable.