Reading a process for its weak points

01Two objects, one name

The documented process is a tidy object. It was written for an audit, a certification or a handover; it lives in a slide or a binder; it describes the work as a clean sequence of steps with clear owners and no waiting. The real process is what happens when an actual order, an actual file or an actual part moves through the organisation on an ordinary day. It contains the waiting, the re-entry of data that already exists somewhere else, the approval that is really a signature hunt, the step performed twice because two departments never agreed which of them performs it once.

The two objects differ in almost every organisation, and the difference is not a failure of documentation. Documentation ages from the day it is published, while the work keeps adapting to reality. The useful point is narrower: only one of the two objects costs money, and it is the one nobody has written down. A review that studies the manual studies the cheaper object. The reading this line performs studies the other one.

02How the real process is read

The reading is simple to describe and demanding to do. Take one order, one file or one part, and follow it end to end at the pace it actually travels, from the moment it enters the process to the moment it leaves as something finished. Not a sample of steps: one unit through all of them, because the truth of a process lives in its sequence, and a sequence cannot be reconstructed from interviews alone.

Along the way, talk to the person who does the step, not the person who owns the slide. The doer knows where the process lies; the owner knows where it is supposed to be honest, and the two kinds of knowledge rarely coincide. And watch where paper waits. Wherever a document, a ticket or a pallet sits untouched, cost is accumulating in silence: cash tied up, commitments ageing, defects travelling downstream undetected. Waiting is the most visible symptom the real process offers and the most commonly ignored, because nothing appears to be going wrong while it happens.

03Where the weak points hide

Weak points rarely sit in the main flow. The main flow is rehearsed daily and has been sanded smooth by repetition. They hide in the exceptions: the special customer, the rush order, the month-end variant, and above all the workaround. Somewhere in the history of almost every process there was one difficult month, a system down, a colleague absent, a deadline immovable, in which someone invented a shortcut to keep the work moving. The month ended. The shortcut stayed. Years later it is still running, unwritten, dependent on one person's memory, and invisible to every review because it appears in no document.

Nobody calls it a weakness. They call it how things are done here. The reader's job is to find these inventions and record them without blame, because the workaround is also information: it marks the exact place where the designed process failed its people, and where the fix should begin.

04The reader's discipline

The discipline of the reading matters as much as its findings. Observation before judgement: a reader who arrives with a diagnosis will find evidence for it, and the people on the floor will recognise the performance and close up. Sequence before summary: the temptation is to jump to totals, to say the process takes eleven days, and to lose the fact that nine of those days are waiting in three specific places. A reading holds its shape when it records what happened, in order, with times and with the names of steps rather than the names of people, and lets the pattern declare itself.

This is slower than opinion. It is also the only version that survives contact with the person who knows the process best, and that person is always in the room when the findings are discussed.

05The deliverable

The deliverable is not a diagnosis. Diagnoses are cheap, and every organisation already holds several, usually contradictory, usually loyal to the department that produced them. The deliverable is a record: this unit entered here, waited there, was touched by these steps in this order, was reworked at this point for this reason, and left after this much time, of which this much was work and this much was waiting.

A record of that precision does something a diagnosis cannot: it survives the meeting where the fix is decided. When the discussion turns political, as it will, the record does not argue back. It simply remains, and decisions made against it hold, because everyone in the room has seen the same process for the first time. That is the quiet product of this line: not advice, but a shared object precise enough to decide on.