Why approval cycles don't get faster when the engineering gets better

The fastest-approved submissions are not always the best-engineered ones. This is one of those observations that experienced design office principals carry quietly for years without ever seeing it stated plainly. You have invested in better tools, tighter drawing standards, more rigorous internal review. The cycle times have not improved in any meaningful way. The reason is that you have been optimising the wrong constraint.

Consider where time actually goes in a typical class submission. The engineering work — calculations, drawing production, internal check — occupies a defined portion of the schedule. That portion has, in fact, improved. What has not improved is everything that happens after the submission leaves your office. Who received it at the classification society, and who should have received it? Is it with the right technical department, or sitting in a general inbox awaiting manual routing? Has the review even started? None of those questions can be answered without a phone call or an email that may itself go unanswered for several days.

The acknowledgment gap alone accounts for more elapsed time than most principals realise. A submission sent on a Monday may not be formally acknowledged until Wednesday. That acknowledgment may be from an administrative contact who has not yet routed the package to the reviewing surveyor. The reviewing surveyor may be on another project. None of this is tracked on your side, because there is nothing to track until a response arrives. The clock runs. The project schedule erodes. The drawing register sits unchanged. Nobody dropped the ball — but nobody is holding it either.

Follow-up is the next sink. Standard practice in most offices is to wait a defined period — two weeks, three weeks, depending on the relationship and the urgency — and then contact the class office to enquire on status. That contact requires knowing who to contact, finding the right person when the original contact is unavailable, and composing a query that is specific enough to be useful. The response, when it comes, often requires a second exchange before status is confirmed. These are not engineering tasks. They require no naval architectural judgment. They do require engineering-grade attention to detail, which is why they tend to fall to the most competent people in the office.

Comment receipt and reconciliation complete the picture. When the class response arrives, it rarely arrives in a format that maps cleanly onto your comment register. Comments reference drawing numbers that may have been superseded. Comments request information that was provided in a different document in the same package. Reconciling the class response against the open comment list is a classification literacy task — it requires someone who understands the submission structure — but it is not engineering design work. It is information management. In most offices it is done by the same engineer who did the calculations, because nobody else has the context to do it.

The administrative layer is structurally invisible for a specific reason: project time is not tracked at that granularity. Hours get logged against the project. They do not get logged against "follow-up to class re: structural package submission" or "comment register update post-approval." The time disappears into general project overhead. When you review a completed project and ask why the approval cycle took eleven weeks, the honest answer is that four of those weeks were absorbed by activities that nobody can precisely account for, because accounting for them was itself one of the tasks that was not being done.

This is not a criticism of classification societies. Their review times are what they are, and they vary by office, surveyor, workload, and the actual complexity of what has been submitted. The point is that the variance you can control is not primarily inside the class office — it is in the handoff, the tracking, and the response management on your side. Better engineering reduces the probability of comments. It does not reduce the elapsed time before you know whether comments are coming.

The offices that have shortened their effective cycle times have not done so by improving their engineering further. They have made the administrative layer visible. They know, at any given moment, the status of every open submission: when it was sent, when it was acknowledged, when review was confirmed to have started, when a response is overdue. They follow up systematically rather than when someone remembers. They log responses the day they arrive. This is not sophisticated project management. It is basic information hygiene applied to a process that most offices have always treated as background noise. The noise turns out to be the schedule.