We run pilots with a small number of companies; the product is not generally available. Publishing per-seat pricing now would be guesswork, and you would be the one paying for the guess. So here is how it actually works today.
How a pilot goes
- 01
Pick one process
Whatever eats the most hours, not the whole system
- 02
Build exactly that
Configured around how you already work
- 03
Run alongside for a month
The old way keeps running, nobody is forced across
- 04
Look at the numbers
The trail shows which stage sped up and which did not
Three things always in the agreement
A ceiling on AI spend
Capped per workspace. At the cap it stops and says so, rather than producing a surprise invoice.
You can stop
A pilot is not a long commitment. If it does not work, stop, and your data stays yours.
The data is yours
Exportable. We do not hold data hostage.
When real pricing arrives
Once the first pilots finish and we know what a process genuinely costs to run. Then the pricing rests on measurements rather than on hope.
The fastest route is a conversation
Tell us where the hours go. If we cannot help, we will say so.