They Asked Us for a Corrective Action Plan. Nobody Had Cited Them for Anything.

A client of ours had a DENR inspection. A few days later his team asked us to prepare the corrective action plan, and to prepare it quickly. We had the plant drawings, we had been through the site, and between us we could have had a credible dated programme written inside a day. I asked one question before we started: what exactly did the inspection cite?

Nobody could answer it. Not because anyone was hiding anything — because the question had not been asked. There was an inspection, there was a visit, there was an atmosphere of something having to be answered, and the assumption underneath all of it was that a document now had to be filed. We did not write the plan that week. This post is about why, and about what we did instead, because the habit we were about to fall into is the single most expensive habit I see on compliance jobs in the Philippines, and it costs nothing to break.

I should be clear about what this is not. It is not a method for keeping things from a regulator. There are things we say in writing even when nobody has asked, and I set them out below — the parts where concealment is the more dangerous option. What this is about is the difference between answering a question and volunteering a confession, which turns out to be a document-design problem rather than an ethical one.

First establish what the event actually is

There are at least four different things a visit from the regulator can be, and they call for four different responses. A Notice of Violation is a citation: it alleges specific things and it has to be answered, and a corrective action plan is the right answer to it. A Cease and Desist Order is a different animal again and the clock on it is short. A technical conference is a conversation you are invited to, usually about something already on record. And a routine inspection cites nothing at all. It is an inspector arriving with a checklist, walking the plant, and leaving.

Our client had the fourth one. No Notice of Violation, no order, nothing alleged in writing. Once that was established the whole shape of the job changed, because the corrective action plan we were two days from writing would have been the first and only document in the file asserting that this facility was non-compliant — and we would have written it ourselves, on the client's letterhead, in answer to nothing.

So the sequence on every compliance job now starts the same way: establish what the event is, get the actual document, and read it before writing a line. Not infer it from how worried everyone sounds. Ask the client for the paper and wait for it. I have got this wrong myself, and I got it wrong in a way worth describing, because it is the ordinary way to get it wrong. Early in that job I had written "client is under a DENR notice" into my own project notes. In the same notes, three pages down, my open-questions list read "the notice as issued — NOV, CDO or routine monitoring." I had recorded that I did not know, and then written as though I did. An item on your own open-questions list is not a fact. If it is still open, the document cannot assume it closed.

An unsolicited plan is a signed admission

A corrective action plan is a good document in its right place. It names defects, it names dates, it names an accountable person. That is exactly what makes it the wrong thing to hand to an inspector who has alleged nothing.

Filed unsolicited after a routine inspection it does three things at once. It states in the client's own words that the facility is not compliant. It fixes dates the client can afterwards be measured and reported against — and a date you set yourself is as binding as one you were given. And it can turn an inspection that was heading for a "for monitoring" outcome into a self-declared violation with a paper trail attached. None of that is the regulator being unfair. It is simply what a document does once it exists.

The posture we settled on for an inspection with no citation is that the programme is prepared and held. It gets written properly, with real dates and real costs, and it stays in the client's folder. It is produced if the inspector raises effluent quality, or asks directly what is being done. Prepared, not confessing. Our client's team found that harder to accept than the engineering, which I understand — doing nothing visible feels like doing nothing. It is not. The work was all happening; it was the filing that stopped.

What we say anyway, because concealing it is worse

Here is where the line sits, and I would rather state it plainly than let the paragraph above stand on its own.

If the plant is not yet compliant, we say so. If the fix will not be finished by the inspection date, we say that too, and we do not let a client imply otherwise. If there is no laboratory result, we say there is none and why — BOD is a five-day incubation and no amount of urgency shortens it, so a result taken this week is not available this week. The same is true of a plant that has only just been started — a new plant can fail its first test for reasons that have nothing to do with how it was built, and the date the sample is taken is a technical decision, not an administrative one. That honesty is protective rather than noble. The worst position a facility can be in is to have claimed a fix that an inspector then goes and looks at. A client who has said "not yet, here is the programme" is in a far stronger position than a client caught overstating. Concealment of a present condition is the one thing that converts a manageable compliance problem into a credibility problem, and credibility is the thing you cannot buy back with equipment.

The same applies to anything we write for a client about their own plant. If their influent has never been tested, or the declared flow on the permit no longer resembles the building's actual flow, or the receiving water has been reclassified under them, or the rules themselves have moved while the plant stood still, those things go to the client in writing, clearly, with what it means and what it costs. We have a page on the compliance side of this that lays out the parameters a facility is actually held to, and the parameters that matter by industry vary enough that it is worth checking which ones apply to you before assuming. A facility can find itself failing without having changed anything, and the class of the water it discharges into decides how much that costs to fix.

A submission answers what was raised and describes future works

When there is a citation and something does have to be filed, the document still does not have to read as a charge sheet.

Take one line from the package we had drafted. "Plant room ventilation inoperative, staff enter daily, danger to life." Compare it with "plant room ventilation renewed." Same work, same scope, same programme, same cost, and no confession. The second version is not less honest than the first — it commits us to exactly the same thing. It just describes what will exist rather than enumerating what does not.

That is the general form. A submission responds to what the notice actually raised, and describes future works. It does not inventory present defects that nobody asked about. Until you know what was cited, you acknowledge only the general position and no specific deficiency, because every deficiency you name is one more thing on the record.

And a drawing is a document. This one took me a while to internalise. Anything printed and hung in the plant room gets read by an inspector who is standing in front of it with a clipboard — the flow diagram, the as-built, the one-line. Every deficiency flag, every "to be replaced", every red cloud we had put there for the client's benefit comes off before it goes on the wall. Show what exists.

Every finding is a separate citation

The reason document count matters more than it looks: findings do not aggregate into one problem. They aggregate into a list of offences.

No discharge permit is its own offence. No accredited pollution control officer is another. An unmaintained logbook is another. An uncontrolled release is another again. Each carries its own penalty. So a nine-finding technical assessment, handed over because it was thorough and because we were proud of it, converts "the inspector found one thing" into "the client admitted nine."

Which is why the questions we now ask before building any regulatory package are about the package and not the plant. Do we need all these documents. Are they all necessary. Is the same information repeated across them. If it is, can one or two carry it. Three documents restating the same position three times is three written admissions, and cutting the repetition cuts the exposure with it. On that job the answer was two documents where we had planned five: a cover letter and the corrective action plan, and only if there was a notice to answer. The technical assessment was reclassified confidential to the client. The terms of reference went to bidders only, offered on request — which gives the client full credit for running an open selection without handing over the yardstick the finished plant will be measured against.

A safety finding moves. It never disappears.

There is one category this logic does not touch, and the distinction is worth stating sharply because it is the easiest place to go wrong.

A life-safety finding is not a compliance line item. Sitting in a regulator's file it becomes a workplace-safety referral and does very little for the person actually at risk. Sitting in the client's hands, addressed to the person who can authorise the work, it becomes an action item with a name on it. So it moves — a separate private letter to the client, marked not for regulatory submission, specific about what and who and when.

It never gets deleted to make a document look tidier. Taking a finding out of a submission and putting it in a letter is document design. Taking it out of both is something else, and if a job ever required that, the honest move is to tell the client we cannot produce the document rather than produce a dishonest one. That is also the test I use on the whole approach: if what we are being asked to write cannot be written without putting the client under the bus, say so to the client instead of releasing the documents. They are usually already under the bus by the time we are called. We do not need to pour petrol on it.

What we showed on the day, and what stayed in the folder

The useful half of that week turned out to be operational rather than clerical. What an inspector takes away from a routine visit is mostly an impression of whether the facility is being run, and that is built from things you cannot produce retrospectively.

So: the logbook open and genuinely being filled, not reconstructed the night before. Disinfection working, with residuals actually logged. The plant room in order. The flow diagram on the wall, showing what exists. The pollution control officer on record. Permits and receipts in one register rather than four drawers. A simple settleability reading taken daily — we use a cone rather than a laboratory number — is worth more in that logbook than a sophisticated test nobody performs. None of that is a paper admission and all of it is visible in ten minutes. It is also, not coincidentally, what a plant being properly operated looks like from the outside, which is why facilities that keep losing their operator tend to inspect badly even when the equipment is sound. Our own maintenance and operation page covers the routine side of it.

There was one more thing worth recording. The programme we wrote and held had a real deadline in it, and it was not a DENR deadline — it was the client's business-permit renewal the following January. That is what made the client move. A plan does not need a regulator holding a copy in order to work; its primary function is internal, and it was already doing that job sitting in a folder. If the parameters driving your own programme are ammonia and phosphate, which is the usual answer — we keep separate notes on ammonia, nitrate and phosphate, and a field guide covering all three — you can get a sense of the cost of the fix from our plant cost estimator or work the sizing through the design generator before committing to dates. Dates you cannot fund are worse than no dates.

Where this does not transfer

Almost all of the above is conditional on circumstances that are not universal, and I would rather name them than let this read as a procedure.

It assumes a routine inspection with nothing cited. With a Notice of Violation in hand the calculus inverts: there is a document to answer, answering it late or thinly is its own problem, and the corrective action plan becomes the right thing to file rather than the wrong one. A Cease and Desist Order removes the discretion entirely. It also assumes a facility whose treatment actually exists in some form — the argument about not enumerating defects has nothing to offer a building discharging raw sewage, where the only real answer is to build the plant, and the choice there is between retrofitting and starting again.

It assumes you have a regulatory relationship to protect over years rather than a single transaction, which is why we take a conservative view of what goes on the record. A client planning to sell the asset next quarter may weigh that differently, and that is their call to make, not ours. Local practice varies more than the national rules suggest — what a regional office does with a self-declared programme is not uniform across the country, and what we see in Cebu is not automatically what you will see elsewhere. Industries under sector-specific regulators, and any facility inside an economic zone with its own environmental unit, operate under reporting duties that may make "prepared and held" simply unavailable to them.

And it assumes somebody is willing to ask the awkward question early, when it is still cheap. The whole of this post is downstream of one question that took thirty seconds to ask and that nobody had asked. Most of what goes wrong on these jobs is like that: not a technical failure but a decision made early, quietly, by assumption. Our own completed work is a reasonable guide to the kinds of facility this experience comes from; the rest of the wastewater side of what we do sits there too, along with what we offer as a treatment company and as a contractor, and how our consultations are arranged.

If you have had an inspection and you are not certain what it was, that is the thing to settle first, and it is a short conversation. Book a free 15-minute consultation.