The pattern: "included" is doing a lot of work in that sentence
A common CCaaS sales pattern lists an AI capability as "included" on a pricing or feature page without specifying the boundary conditions: is it included for every agent, every channel, unlimited volume? In practice, "included" AI features are frequently included at a starter volume (a fixed number of AI-assisted interactions per month) with overage billed separately, included only for a subset of channels (voice AI but not chat AI, for example), or included as a stripped-down version of the capability, with the fuller version reserved for a higher tier.
Why this happens: AI usage has real marginal cost
This is not necessarily a bait-and-switch — AI inference has real, ongoing compute cost that scales with usage in a way that a static software feature (like a reporting dashboard) does not. A vendor genuinely cannot offer unlimited AI usage at a fixed price without either building that marginal cost into every seat’s base price (raising the sticker price for everyone, including light AI users) or metering it separately for users who consume more. The problem is not that AI usage gets metered — it is when that metering is not disclosed clearly at the point a buyer is evaluating "included" features.
What to ask in an RFP to surface this before signing
For every AI feature listed as included: ask for the specific volume limit (interactions per month, minutes, sessions) at which overage billing begins, ask what the overage rate is, ask whether the limit applies per seat or as a shared organizational pool, and ask whether the "included" version is functionally identical to the capability demonstrated in the sales process or a reduced version of it. Get the answers in writing, not verbally in a sales call, since these terms belong in the contract, not just the pitch.
How to model this into a real cost comparison
When comparing vendors, build the AI usage line item into total cost of ownership using your own projected volume, not the vendor’s advertised starter allotment. A platform with a lower headline per-seat price but a low AI usage cap and steep overage pricing can cost more at your actual volume than a platform with a higher headline price and unlimited or more generously capped AI usage. This is the same discipline described in the companion piece on measuring AI ROI: use your own numbers, not a vendor’s framing.
Applying this discipline to Voz360 specifically
A buyer evaluating Voz360 should apply the identical scrutiny: ask which tier includes Answer Engine and Context Retrieval, whether either has a usage cap, and what a scoped quote would look like at actual projected interaction volume, since Voz360’s published pricing page is an illustrative tier structure with exact figures pending business sign-off. The point of this article is the evaluation discipline, not a claim that Voz360’s eventual pricing will avoid this pattern — that should be verified directly, the same as with any vendor.
Can the vendor tell you — in one sentence — which of their AI capabilities are rule-based, which are generative, and which are still roadmap?