Named-seat pricing: one license per person
Under named-seat pricing, each individual agent who might ever use the system needs their own licensed seat, tied to that specific person’s identity. If an organization has 100 agents across three shifts but never more than 40 working at the exact same time, named-seat pricing still requires paying for all 100 licenses, because the cost model is based on total licensed users, not simultaneous usage.
Concurrent-agent pricing: paying for peak simultaneous usage
Under concurrent-agent pricing, the cost is based on the maximum number of agents logged in and actively working at the same moment, not the total number of individuals who might ever log in. The same 100-agent, three-shift organization with a peak of 40 simultaneous agents would size its concurrent-agent license count around that 40-agent peak, potentially at a lower total cost than licensing all 100 named seats, depending on the per-unit pricing of each model.
Why the right choice depends on shift structure and usage pattern
Concurrent-agent pricing tends to favor organizations running multiple shifts, part-time or rotating staff, or seasonal workforce scaling, where the total number of people who need occasional access meaningfully exceeds the number working at any single moment. Named-seat pricing tends to be simpler to forecast and can be more cost-effective for an organization where most licensed agents work roughly the same hours and the gap between total headcount and simultaneous peak usage is small.
What to watch for when comparing quotes across these two models
A quote based on named seats and a quote based on concurrent agents are not directly comparable by their headline per-unit price alone — a lower per-concurrent-agent price could still cost more in total than a higher per-named-seat price, depending on an organization’s actual shift overlap and peak usage pattern. Model the actual cost under both pricing structures using real staffing and shift data before comparing vendors, rather than comparing sticker prices across two different pricing units.
Can the vendor tell you — in one sentence — which of their AI capabilities are rule-based, which are generative, and which are still roadmap?