Pricing is shaped by scope, sensitivity, deployment model, and how much rollout support is needed. Exact figures come after discovery — the ranges below are directional.
Engagement models
Best for teams that need a serious architecture and readiness recommendation before funding a pilot.
Best for a first use case where the organization wants a controlled deployment with defined success criteria.
Best when a validated pilot is ready to move into a production environment with broader adoption.
Best for teams that want continued advisory, policy review, optimization, and expansion planning.
Ranges reflect the scope a typical mid-market to enterprise client funds — they are not minimums, maximums, or a substitute for a quote.
What drives the quote
A real quote depends on these. They are surfaced during discovery so the proposal is grounded in your actual environment.
Higher sensitivity means more rigorous control design, deeper threat modeling, and tighter logging boundaries — all of which add scope.
On-premises and air-gapped environments cost more than private cloud, which costs more than controlled hosted. The choice is driven by your constraints, not by us.
Each system the AI environment reads from or writes to adds review, access, and operational complexity. A clean pilot touches 1–3 systems; rollouts often touch 5+.
One team behind one workflow is the cheapest engagement. Multiple roles, geographies, or business units multiply the design and training surface.
If access reviews, change control, and logging review already exist internally, governance work plugs in faster. If those processes need to be built, that effort is part of the engagement.
Light support is documentation plus a launch review. Deeper support includes training cohorts, office hours, and an extended hardening window.
What is intentionally not part of this pricing
Exact pricing belongs after a discovery conversation. That keeps the proposal grounded in environment constraints, ownership model, and the level of governance required.
