Permitting goes better when you understand what the person reading your submittal is trying to do.
They are not trying to stop your project. They are trying to reach a decision they can defend — to their supervisor, to a commenting agency, to the public, and potentially to a court. Everything that helps them do that speeds you up.
What they are actually assessing
Different programs test different things, but the recurring questions are:
Is the problem real, and is it characterised? What is eroding, why, how fast, and what happens if nothing is done.
Was avoidance considered? Is there a way to solve this with less impact, including a non-structural option? For Section 404 this is formally the first step in the sequence.
Is the impact minimised? Footprint, duration, method, timing. Is the proposed work proportionate to the problem?
What are the effects, and on what? Aquatic habitat, hydraulics, adjacent properties, navigation, public access, listed species, cultural resources.
Is it going to work? Not their primary role, but a reviewer who thinks the design will fail in three years is being asked to authorise a repeat application.
Is it consistent with the program’s policies? The enforceable policies are published, and the determination tests against them.
Can the decision be defended? The quiet one, and often the real constraint.
What makes a package easy
A clear problem statement, with evidence. Photographs over time, survey, measured loss rates. A reviewer who can see the problem does not have to take it on trust.
A genuine alternatives discussion. Including alternatives you rejected and why. A submittal that names only the proposed work has skipped the first two steps of the sequence and reads as though the answer preceded the analysis.
Minimisation stated explicitly. Do not leave it implicit. If the design avoids dewatering, says so and says why that matters. If the footprint follows the existing bank line rather than projecting into the channel, say so. If placement is unit by unit into an existing hole rather than grading the bed flat first, say so. These are minimisation arguments and reviewers do not infer them.
Drawings that answer the questions being asked. Plan and section, existing and proposed bed elevations, the ordinary high water mark, extents, and the relationship to the channel or navigation line. See assembling the submittal package.
Documentation for the materials. What the material is, what backs the claims, and where it comes from.
Consultations already started. A package arriving with species and cultural resource work underway is much further along than one that triggers them at intake.
Realistic timing that respects the work window.
The thing that helps most, and is done least
State the limitations yourself.
A reviewer’s job is to find the gaps. If your submittal has already identified them, two things happen: the review is faster because they are not hunting, and the rest of the package becomes more credible because you have demonstrated you are not hiding anything.
This is why we publish when not to use rock bags — a page built from the questions our own data does not answer, including that we hold no stability data for some conditions, no scour-reduction figure, and no citable model tests for others. It exists to be quoted by someone assembling a package, and by a reviewer testing it.
It also means we publish which test method backs which claim rather than measured values, and that is worth stating in a submittal rather than leaving to be discovered. A reviewer who needs a measured number will ask; better they ask in week one.
What slows a review
An incomplete package. The single biggest cause. The clock runs from complete, and information requests reset the round trip.
A footprint larger than the problem. It invites the question of whether the whole thing is necessary, which reopens the alternatives analysis.
New hard structure where a softer option would serve, particularly in a coastal program with policies against it. See state coastal zone approvals.
No evidence the problem exists. “The bank is eroding” without measurement.
Claims that cannot be supported. A performance number with no source behind it is worse than no number, because now the reviewer distrusts the rest.
Consultations started at submission. They should be running already. See Endangered Species Act consultation.
After the fact. Work already placed without authorisation changes the conversation entirely, from permitting to enforcement.
Ignoring the neighbours. End effects on adjacent property are a legitimate concern and they generate comments.
Practical suggestions
Have a pre-application meeting. Usually free, usually offered, and it tells you what this district or this state actually wants. It is the highest-return hour in the whole process.
Write for someone with a large caseload. Lead with the problem and the proposal. Put the supporting material behind it.
Answer the questions the program asks, in the program’s own vocabulary.
Be responsive. Information requests answered in days rather than weeks compress the elapsed time far more than anything in the original submission.
Do not argue with a condition you can live with. Conditions are how a reviewer gets to yes.
Where to go next
- Assembling the submittal package for the contents
- Permitting in-water erosion work for the overview
- Section 404 for the avoid-minimise-mitigate sequence
- Getting a non-standard product approved for materials approval
- When not to use rock bags for the limitations, written to be quoted
- The permitting hub for the rest