The questions every buyer asks
Frequently asked questions
Q:What does Managentics do?
We provide a platform for agentic process specification: we specify, govern and audit organizations' agentic AI processes. We do not operate runtimes, which keeps our assessment independent.
Q:Who owns the results?
The client. The specifications, process descriptions, work instructions and governance documents are the client's property and travel with the client from vendor to vendor.
Q:What technology do we commit to?
None. The specification is technology-independent, and the implementation platform can be chosen and replaced without the specification suffering.
Q:Where do the processes run and where is the data?
In the client's own cloud environment, on the client's own accounts. The specification platform handles process descriptions, not production data.
Q:How is pricing structured?
The specification phase is fixed-price, so you know the cost before starting. Continuation is agreed stage by stage, and no stage commits you to the next. We give an exact offer after the discovery call.
Q:Can our own IT team implement the processes?
Yes, and the model is built for it. The target state is that your own small team builds the processes from the specifications, with us supporting through the platform and technical consulting.
Q:How are the human role and responsibility ensured?
Approval points are designed into the specification before implementation, and every approval leaves a trail at runtime. Responsibility does not shift to the machine by accident.
Q:How does the EU AI Act affect us?
It depends on the risk class of your systems. The control points, risk classifications and run trail recorded in the specification form a foundation on which the compliance assessment is light to do.
Q:What if the cooperation with Managentics ends?
All specification and documentation remain with you, and the runtime is in your own hands. Continuity does not depend on us, and that is by design the core property of the model.
Q:Who is the service for?
Best suited to expert organizations with recurring, knowledge-intensive processes where the correctness of outcomes matters. If your need is one light automation, we will say directly if an off-the-shelf product is the better choice.
Q:Do we get locked into one vendor or platform?
No. The specification is technology-independent and the client's property, so the same description can be implemented on another platform or by another implementer without redoing the specification work. We do not implement or operate runtimes ourselves, so we have no incentive to lock you in.
Q:How do we know what the agent is really doing?
Every run leaves an inspectable trail: which steps ran, on what data, what the agent decided, who approved and where the run stopped. The trail is compared against the specification, so deviations are visible instead of hidden inside the model's internal logic.
Q:How do token costs stay under control?
Token costs come from unnecessary runs, oversized context, over-specified models and loops that repeat the same work. The specification decides which step runs when, what information is passed to the model, which model is sufficient for that step and on what condition a run is cut off. The run trail reports cost step by step, so the expensive places are visible before the invoice grows.