Bring us meaningful work.
One repository. One delivery path. Rules and measures agreed before anything runs, and a written recommendation at the end to continue, narrow or stop. The ask is deliberately small, and you can end it.
Reign Factory is in beta and being productized. It is not generally available; beta access is agreed with us and scoped in writing before anything runs. Deployment is a single-tenant dedicated instance.
Real work, not a demonstration repository.
Bring something that is already on someone's list and already costing you something.
Real work carries the constraints that decide whether a delivery path holds: a review path with a person at the end of it, a release window that belongs to someone else, a dependency nobody wants to touch, a rule that is written down and enforced. Those constraints are the test. Everything else is theatre.
The work is understood, the fix is agreed, and nobody has had a clear run at it. This shape suits us because done has already been defined by someone else.
AskWhich of these items has been reopened after being closed?
The work is repetitive, the risk sits in the review, and the review stays with you.
AskWhat is the oldest thing you are still paying to keep alive?
Bring the part that is blocking it rather than the whole program.
AskIf one piece of this moved first, which piece would it be?
We work inside your repository, under your controls, on tooling we operate, including GitLab. A named engineer embedded for this population runs the engagement alongside your reviewer.
What has to be true before we start.
An honest checklist. If an item is missing, we would rather find out now. These four are the whole entry requirement; there is nothing to install on your side and nothing to procure in order to have the conversation.
A repository we can be given access to
Under your controls. Your identity provider, your permissions, your network rules. We do not ask you to relax any of them, and we do not take a copy elsewhere.
A definition of what done looks like
Written by you, and specific enough that a reviewer can say yes or no to it without calling a meeting.
A named reviewer on your side
One person who is accountable for accepting or rejecting the work, and who has the authority to do it.
Agreement on what will be measured
The baseline is recorded before the first change, not reconstructed afterwards. If we cannot agree a measure, we do not claim one.
Work, measures, and a recommendation.
Three things, at the end, in writing.
The three things
Completed work in your own approval path. Changes arrive for review in the repository you gave us. Your named reviewer approves or rejects. We do not merge and we do not release.
Measures against the baseline you agreed. Reported against the definition set before the work started, with the record of what was done and by what path.
A written recommendation to continue, narrow or stop. With the reasoning, so your own people can argue with it.
Where it stops
Stop is a real outcome. If the evidence points there, we will write it down and say why. A recommendation that can only ever say continue is not worth reading, and it is not what we are offering.
Our figures are ours. The only figures we publish are measured on our own engineering, in our own internal environment. That is our work in our environment. It is not a projection for your estate and we will not present it as one.
Nothing here is a purchase. There is no agreement, no statement of work and no commercial terms in place by reading this page or by writing to us.
The method behind the published figures, including what was counted and what was excluded, is written up in full. Read the method →
Send us one repository.
One repository, one definition of done, one named reviewer. If the answer at the end is stop, you will have that in writing and you will have lost very little getting it.