The conditions not in the job description¶
A systems architect designs and oversees complex IT systems, ensuring that hardware, software, and networks work together to meet organisational goals. That description is accurate as far as it goes. It describes the rational layer of the work: the technical activities, the outputs, the deliverables. It does not describe the conditions under which that work succeeds or fails, and it does not describe the two other layers that determine whether an architecture produces the outcomes it was designed to produce.
The model problem¶
A systems architect is, in the SEM sense, a person whose primary product is a model: a representation of what the system is, how it works, and what it is for. Models encode assumptions. Assumptions drift. The most technically accomplished architecture documentation in the organisation is also, over time, one of the most reliably inaccurate representations of the actual system. Managing that gap is as much the architect’s work as producing the original design.
The job description does not mention this. It describes the creation of architectures, not the ongoing responsibility for their accuracy. An architect who understands their role as producing designs rather than maintaining models will produce increasingly outdated documentation and will not understand why the systems they designed keep surprising them.
The change problem¶
Architectural decisions are interventions in systems that are already in steady state. Every significant architectural change disrupts existing processes, roles, and investments. That disruption is not an obstacle to the work; it is part of the work. An architect who designs without attending to the homeostatic resistance of the organisation will produce excellent designs that are not implemented, or that are implemented in ways that preserve the old patterns under new labels.
The job description does not mention this either. It describes collaboration with stakeholders, which is the visible surface of what is actually required: understanding the existing steady state well enough to design change that can take root in it, and creating the conditions for that change to be sustained rather than reversed.
The social problem¶
Architecture work is irreducibly social. The architect who attends to the rational layer but not the emotional and political layers will find that technically correct decisions produce unexpected outcomes: teams that nominally comply but practically route around the new architecture, stakeholders who agreed in the room and quietly resumed the prior approach, review processes that surface concerns too late for them to be addressed.
The job description mentions stakeholder collaboration and business alignment. It does not describe what that work actually requires: attending to people’s investments in the existing system, understanding whose authority sanctions architectural decisions, and creating the conditions for honest engagement rather than compliance theatre.
What the foundations add¶
None of this makes the technical work less important. It makes the technical work contingent on conditions that are not themselves technical, and that are not described in any job description or competence framework for the role.
The SEM framing gives architects a way to understand their documentation as a living model with a maintenance responsibility, not a deliverable with a completion date. The ChangeShop framing gives architects a way to understand organisational resistance as information about the change they are asking the organisation to make. The PSL framing gives architects a way to see that their work operates simultaneously at rational, emotional, and political layers, and that attending to all three is not optional. The Satir framing gives architects a way to understand the communication patterns that produce architectural dysfunction, and the conditions under which honest architectural review becomes possible.
These are not additions to the technical role. They are the conditions under which the technical role produces what it is supposed to produce.