How per-seat pricing works, and where it fits
Per-seat pricing charges a fixed amount per licensed agent per billing period, independent of call volume, message volume, or AI usage within that seat’s normal use. It is the dominant model in enterprise CCaaS because it is easy to budget and easy to forecast: headcount planning and license cost planning become the same exercise. It fits organizations with relatively stable, predictable interaction volume per agent — the seat cost reflects a consistent unit of capacity.
How consumption pricing works, and where it fits
Consumption pricing charges based on a usage unit — per session, per minute, per resolved conversation, per AI-assisted interaction — rather than per license. It fits organizations with highly variable volume (seasonal spikes, unpredictable campaign-driven outbound bursts) where per-seat pricing would mean paying for idle capacity during slow periods, or paying to provision extra seats ahead of a spike. The tradeoff is forecasting difficulty: a consumption bill is harder to predict precisely than a fixed seat count, and a usage spike (intentional or not) directly increases cost in a way per-seat pricing does not.
The hybrid reality: most vendors blend both
Very few CCaaS platforms are purely one model. A common pattern is per-seat pricing for the base platform (voice, digital channels, core workflow) with consumption-based add-ons for specific high-variable-cost features — AI usage in particular is frequently metered even when the base platform is sold per-seat, because AI inference cost scales with usage in a way that voice-seat capacity does not. A buyer evaluating "per-seat pricing" should confirm which specific capabilities are actually included in the seat price and which are metered separately.
What to ask in an RFP regardless of model
Ask for the full unit economics: what exactly is included in the base seat or base usage unit, what triggers an overage charge, how overages are billed (hard cap, automatic upgrade, or pay-as-you-go), and whether AI features specifically are seat-included or consumption-metered. A quoted per-seat number that excludes metered AI usage is not comparable to a competitor’s per-seat number that includes it, even if the headline figures look similar.
Voz360’s illustrative tier structure as one example
Voz360’s published pricing page uses an illustrative, feature-mapped tier structure (Core, Omnichannel, Enterprise) rather than a granular consumption meter, with specific dollar figures pending business sign-off. It is shown here only as one example of how a tiered, feature-inclusive structure can be organized — not as a claim that this structure is finalized or that it represents the only viable pricing model for the category. A buyer comparing Voz360 to any other vendor should request the same unit-economics detail described above rather than comparing tier names alone.
Can the vendor tell you — in one sentence — which of their AI capabilities are rule-based, which are generative, and which are still roadmap?