Template:M intro design process

Integrity of the process is everything in modern risk management dogma. Hence risk taxonomies, service catalogs and playbooks.

Following an algorithm requires no comprehension of the content of the process — you don’t need to know how an internal combustion engine works to drive a car — and comprehension even risks subversion of the process: if subject matter expertise might incline one to take a view on process step, such that the form breaks, even though the substance does not, and if your means of control is predicated on the immaculate operation of process, there is something to be said for removing subject matter experts from the process altogether, and staffing it with those with no insight, who will be inclined to be as machine-like as can be. The process is its overlord.

Wisdom, judgment, pragmatism, “metis”: these might accelerate a particular item through the system, but at a cost to the integrity of the process.

Holding this world view one might appoint business controllers with no subject matter understanding, let alone metis, at all. They administrate, rather than manage. To a process aficionado, one who suddenly asks his legal eagles to prove that one can terminate a standard stock loan on notice, as if he can scarcely believe it — quite a specific example, I know — is demonstrating an impressively fearless readiness to challenge conventional wisdom — an almost child-like, savant curiosity. Once upon a time, such a person would have had no business being employed in financial services at all.

Subject matter experts are also expensive. This is a cardinal sin in an industry where the highest calling is cost reduction.

The ideal “process participant” costs nothing, follows instructions with perfect fidelity, doesn’t break down, makes no errors, and certainly doesn’t think or question the process: that is, it is a computer. In the same way a machine doesn’t question its program (it can’t), a process participant escalates within the process, but doesn’t question it. the difference is that cantankerous human process participants can.

But therein the problem: if the process can be computerised, why hasn’t it been?

There is a paradox here, though, because to get the best outcome within the playbook parameters requires a degree of advocacy, inasmuch as the process participant is facing the outside world (beyond the playbook control) - you can best negotiate if you understand your subject material.

The portfolio risk engine ascribes the same value to any outcome as long as it conforms to the playbook. The principle measurement is cost (lack of) and then speed.

The theory is we operationalise a negotiation process. We divide into doers — process participants and thinkers “process designers”. Wherever there is a playbook, the demands of fidelity and economy require a deskilling and de-emphasis of subject matter expertise from the process participants.

The same does not hold for the process designers. BUT — and here's the thing: if we also operationalise the escalation process — and the dogma of internal audit and the bottom line imperative see to it that we do — we wind up with a series of nested playbooks stretching up and across the organisation, and the real expertise (internal audits) becomes expertise in the operational parameters of the different layers and abstractions of operational playbook: reconciling them, testing them for consistency and compatibility, while in the mean time subject matter expertise — of the actual substantive content of the operation — has leaked out of the whole system.